Load the following libraries. If they are not installed, run install.packages (“packagename”)
#from https://www.vikram-baliga.com/blog/2015/7/19/a-hassle-free-way-to-verify-that-r-packages-are-installed-and-loaded
packages = c("tidyverse","ggplot2","beeswarm","colorspace", "stargazer", "qwraps2", "gridExtra","ggpubr","car","olsrr","ggbeeswarm", "broom", "rmarkdown", "table1", "kableExtra", "epitools", "FSA", "Hmisc", "ufs")
package.check <- lapply(packages, FUN = function(x) {
if (!require(x, character.only = TRUE)) {
install.packages(x, dependencies = TRUE)
library(x, character.only = TRUE)
}
})
#verify they are loaded
search()
First, import the data from the data (CSV) file
#setwd("~/etc") #This should be wherever your file is saved
BRIEF <- read.csv("SLaM_BRIEF-P raw data 20200911.csv", na.strings = c("N/A", "", "Unknown", "Excluded"))
BRIEF <- subset(BRIEF, BRIEF$AgeMonths < 100 & BRIEF$AgeMonths!="" & BRIEF$SES..3.66.!="" & BRIEF$UnilateralHearingLoss. == "No")
BRIEF$Language_Modality <- factor(ifelse(BRIEF$Group_4cat == "English Early" | BRIEF$Group_4cat == "English Later", "English", "ASL"), levels = c("English", "ASL"))
BRIEF$Language_Timing <- factor(as.character(BRIEF$Group_2cat), levels = c("Early", "Later"), exclude="")
BRIEF$LanguageGroup <- as.factor(factor(as.character(BRIEF$Group_4cat), levels = c("English Early", "ASL Early", "English Later", "ASL Later"), labels = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), exclude=NA))
BRIEF$Hearing_Level_3_Cat <- as.factor(factor(as.character(BRIEF$Hearing_Level_3_Cat), levels = c("TH", "deaf", "HoH"), labels = c("Typically Hearing", "Deaf", "Hard of Hearing"), exclude=NA))
BRIEF$Hearing.Device <- dplyr::recode(BRIEF$Hearing.Device, "BAHA" = "Other")
BRIEF$Hearing.Device <- as.factor(factor(as.character(BRIEF$Hearing.Device), levels = c("HA", "CI", "HA & CI", "Other", "None"), labels = c("Hearing Aid", "Cochlear Implant", "Hearing Aid & Cochlear Implant", "Other", "None"), exclude=NA))
#Recode Race to have fewer categories
BRIEF$Race_recoded <- dplyr::recode(as.character(BRIEF$Race), 'Asian' = "Asian", 'Black or African American'="Black or African American", 'More than One'="More than one", 'Other'="Other/Missing", 'Unsure, or prefer not to answer' = "Other/Missing", 'White'="White", .missing="Other/Missing")
#update ORDER of groups
BRIEF$Race_recoded <- factor(BRIEF$Race_recoded, levels = c("Asian", "Black or African American", "White", "More than one", "Other/Missing"), labels = c("Asian", "Black or African American", "White", "More than one", "Other/Missing"))
#Recode Ethnicity to have fewer categories
BRIEF$Ethnicity_recoded <- dplyr::recode(as.character(BRIEF$Ethnicity), 'Hispanic' = "Hispanic", 'NonHispanic'="Non-Hispanic", 'PreferNotToAnswer'="Prefer not to answer", .missing="Missing")
#update ORDER of groups
BRIEF$Ethnicity_recoded <-factor(BRIEF$Ethnicity_recoded, levels = c("Hispanic", "Non-Hispanic", "Prefer not to answer", "Missing"), labels = c("Hispanic", "Non-Hispanic", "Prefer not to answer", "Missing"))
table1::label(BRIEF$AgeMonths) <- "Age (Months)"
table1::label(BRIEF$SES..3.66.) <- "SES"
table1::label(BRIEF$Sex) <- "Sex"
table1::label(BRIEF$Race_recoded) <- "Race"
table1::label(BRIEF$Ethnicity_recoded) <- "Ethnicity"
table1::table1(~AgeMonths + SES..3.66. + Sex + Race_recoded + Ethnicity_recoded, data = BRIEF)
| Overall (N=123) |
|
|---|---|
| Age (Months) | |
| Mean (SD) | 60.1 (13.5) |
| Median [Min, Max] | 59.0 [37.0, 91.0] |
| SES | |
| Mean (SD) | 49.8 (14.7) |
| Median [Min, Max] | 54.0 [8.00, 66.0] |
| Sex | |
| Female | 66 (53.7%) |
| Male | 57 (46.3%) |
| Race | |
| Asian | 3 (2.4%) |
| Black or African American | 1 (0.8%) |
| White | 104 (84.6%) |
| More than one | 10 (8.1%) |
| Other/Missing | 5 (4.1%) |
| Ethnicity | |
| Hispanic | 10 (8.1%) |
| Non-Hispanic | 97 (78.9%) |
| Prefer not to answer | 1 (0.8%) |
| Missing | 15 (12.2%) |
table1::table1(~AgeMonths + SES..3.66. + Sex + Race_recoded + Ethnicity_recoded | LanguageGroup, data = BRIEF, overall=F)
| Typically Hearing (N=46) |
Early ASL (N=26) |
Later English (N=23) |
Later ASL (N=28) |
|
|---|---|---|---|---|
| Age (Months) | ||||
| Mean (SD) | 54.7 (10.9) | 60.8 (14.4) | 60.7 (10.9) | 68.0 (14.6) |
| Median [Min, Max] | 54.5 [37.0, 85.0] | 57.0 [41.0, 91.0] | 60.0 [37.0, 78.0] | 71.5 [37.0, 90.0] |
| SES | ||||
| Mean (SD) | 55.6 (8.83) | 48.2 (16.8) | 48.8 (14.1) | 42.7 (17.4) |
| Median [Min, Max] | 56.0 [21.5, 66.0] | 56.5 [11.0, 66.0] | 51.0 [8.00, 66.0] | 49.0 [9.00, 62.0] |
| Sex | ||||
| Female | 24 (52.2%) | 16 (61.5%) | 13 (56.5%) | 13 (46.4%) |
| Male | 22 (47.8%) | 10 (38.5%) | 10 (43.5%) | 15 (53.6%) |
| Race | ||||
| Asian | 0 (0%) | 0 (0%) | 0 (0%) | 3 (10.7%) |
| Black or African American | 0 (0%) | 0 (0%) | 1 (4.3%) | 0 (0%) |
| White | 43 (93.5%) | 24 (92.3%) | 18 (78.3%) | 19 (67.9%) |
| More than one | 3 (6.5%) | 1 (3.8%) | 3 (13.0%) | 3 (10.7%) |
| Other/Missing | 0 (0%) | 1 (3.8%) | 1 (4.3%) | 3 (10.7%) |
| Ethnicity | ||||
| Hispanic | 3 (6.5%) | 0 (0%) | 2 (8.7%) | 5 (17.9%) |
| Non-Hispanic | 42 (91.3%) | 18 (69.2%) | 17 (73.9%) | 20 (71.4%) |
| Prefer not to answer | 0 (0%) | 1 (3.8%) | 0 (0%) | 0 (0%) |
| Missing | 1 (2.2%) | 7 (26.9%) | 4 (17.4%) | 3 (10.7%) |
ggboxplot(BRIEF, x = "LanguageGroup", y = "SES..3.66.",
color = "LanguageGroup",
order = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"),
ylab = "SES", xlab = "Language Group")
ggplot(data=BRIEF, mapping=aes(x=SES..3.66.))+ geom_histogram(binwidth=10) + facet_grid(~LanguageGroup)
leveneTest(SES..3.66.~LanguageGroup, data=BRIEF)
## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 3 3.6537 0.01457 *
## 119
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
SES_LangGrp <- kruskal.test(SES..3.66.~LanguageGroup, data=BRIEF)
SES_LangGrp
##
## Kruskal-Wallis rank sum test
##
## data: SES..3.66. by LanguageGroup
## Kruskal-Wallis chi-squared = 11.705, df = 3, p-value = 0.008465
dunnTest(SES..3.66.~LanguageGroup, data=BRIEF, method="hochberg")
## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Hochberg method.
## Comparison Z P.unadj P.adj
## 1 Early ASL - Later ASL 1.712379 0.0868268640 0.347307456
## 2 Early ASL - Later English 0.616999 0.5372354018 0.537235402
## 3 Later ASL - Later English -1.029644 0.3031773783 0.606354757
## 4 Early ASL - Typically Hearing -1.334535 0.1820284853 0.546085456
## 5 Later ASL - Typically Hearing -3.311757 0.0009271197 0.005562718
## 6 Later English - Typically Hearing -1.973772 0.0484076294 0.242038147
#selected hochberg correction method referencing this site: https://towardsdatascience.com/an-overview-of-methods-to-address-the-multiple-comparison-problem-310427b3ba92
Findings:
ggboxplot(BRIEF, x = "LanguageGroup", y = "AgeMonths",
color = "LanguageGroup",
order = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"),
ylab = "Age (Months)", xlab = "Language Group")
ggplot(data=BRIEF, mapping=aes(x=AgeMonths))+ geom_histogram(binwidth=6) + facet_grid(~LanguageGroup)
leveneTest(AgeMonths~LanguageGroup, data=BRIEF)
## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 3 1.255 0.2931
## 119
Age_LangGrp <- kruskal.test(AgeMonths~LanguageGroup, data=BRIEF)
Age_LangGrp
##
## Kruskal-Wallis rank sum test
##
## data: AgeMonths by LanguageGroup
## Kruskal-Wallis chi-squared = 16.981, df = 3, p-value = 0.000713
dunnTest(AgeMonths~LanguageGroup, data=BRIEF, method="hochberg")
## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Hochberg method.
## Comparison Z P.unadj P.adj
## 1 Early ASL - Later ASL -2.0851898 3.705207e-02 0.1852603675
## 2 Early ASL - Later English -0.3394709 7.342550e-01 0.7342550228
## 3 Later ASL - Later English 1.6727536 9.437580e-02 0.2831274122
## 4 Early ASL - Typically Hearing 1.6716134 9.460059e-02 0.1892011775
## 5 Later ASL - Typically Hearing 4.0804060 4.495711e-05 0.0002697427
## 6 Later English - Typically Hearing 1.9865461 4.697272e-02 0.1878908797
Findings:
Approach: Compare (via Welch’s t-tests) the BRIEF raw scores of Early English group to those of Early ASL group
Create dataframe with only children exposed to language early and perform t-tests for all BRIEF-P scales
BRIEF_early <- subset(BRIEF, BRIEF$Language_Timing=="Early")
table1::label(BRIEF_early$AgeMonths) <- "Age (Months)"
table1::label(BRIEF_early$SES..3.66.) <- "SES"
table1::label(BRIEF_early$Sex) <- "Sex"
table1::label(BRIEF_early$Race_recoded) <- "Race"
table1::label(BRIEF_early$Ethnicity_recoded) <- "Ethnicity"
table1::table1(~AgeMonths + SES..3.66. + Sex + Race_recoded + Ethnicity_recoded, data = BRIEF_early)
| Overall (N=72) |
|
|---|---|
| Age (Months) | |
| Mean (SD) | 56.9 (12.5) |
| Median [Min, Max] | 55.0 [37.0, 91.0] |
| SES | |
| Mean (SD) | 52.9 (12.7) |
| Median [Min, Max] | 56.0 [11.0, 66.0] |
| Sex | |
| Female | 40 (55.6%) |
| Male | 32 (44.4%) |
| Race | |
| Asian | 0 (0%) |
| Black or African American | 0 (0%) |
| White | 67 (93.1%) |
| More than one | 4 (5.6%) |
| Other/Missing | 1 (1.4%) |
| Ethnicity | |
| Hispanic | 3 (4.2%) |
| Non-Hispanic | 60 (83.3%) |
| Prefer not to answer | 1 (1.4%) |
| Missing | 8 (11.1%) |
table1::table1(~AgeMonths + SES..3.66. + Sex + Race_recoded + Ethnicity_recoded | LanguageGroup, data = BRIEF_early, overall=F)
| Typically Hearing (N=46) |
Early ASL (N=26) |
|
|---|---|---|
| Age (Months) | ||
| Mean (SD) | 54.7 (10.9) | 60.8 (14.4) |
| Median [Min, Max] | 54.5 [37.0, 85.0] | 57.0 [41.0, 91.0] |
| SES | ||
| Mean (SD) | 55.6 (8.83) | 48.2 (16.8) |
| Median [Min, Max] | 56.0 [21.5, 66.0] | 56.5 [11.0, 66.0] |
| Sex | ||
| Female | 24 (52.2%) | 16 (61.5%) |
| Male | 22 (47.8%) | 10 (38.5%) |
| Race | ||
| Asian | 0 (0%) | 0 (0%) |
| Black or African American | 0 (0%) | 0 (0%) |
| White | 43 (93.5%) | 24 (92.3%) |
| More than one | 3 (6.5%) | 1 (3.8%) |
| Other/Missing | 0 (0%) | 1 (3.8%) |
| Ethnicity | ||
| Hispanic | 3 (6.5%) | 0 (0%) |
| Non-Hispanic | 42 (91.3%) | 18 (69.2%) |
| Prefer not to answer | 0 (0%) | 1 (3.8%) |
| Missing | 1 (2.2%) | 7 (26.9%) |
BRIEF Scores & Welch two sample t-tests for two “Early” groups
table1::label(BRIEF_early$GEC_RawScore) <- "Global Executive Composite"
table1::label(BRIEF_early$Inhibit_RawScore) <- "Inhibition"
table1::label(BRIEF_early$Shift_RawScore) <- "Shift"
table1::label(BRIEF_early$Emotional.Control_RawScore) <- "Emotional Control"
table1::label(BRIEF_early$Working.Memory_RawScore) <- "Working Memory"
table1::label(BRIEF_early$Plan.Organize_RawScore) <- "Plan/Organize"
table1(~GEC_RawScore + Inhibit_RawScore + Shift_RawScore + Emotional.Control_RawScore + Working.Memory_RawScore + Plan.Organize_RawScore | Language_Modality, data = BRIEF_early, overall=F)
| English (N=46) |
ASL (N=26) |
|
|---|---|---|
| Global Executive Composite | ||
| Mean (SD) | 87.8 (18.7) | 89.5 (16.1) |
| Median [Min, Max] | 87.5 [63.0, 155] | 90.0 [63.0, 132] |
| Inhibition | ||
| Mean (SD) | 22.8 (5.93) | 24.0 (6.06) |
| Median [Min, Max] | 23.0 [16.0, 46.0] | 22.5 [16.0, 37.0] |
| Shift | ||
| Mean (SD) | 13.0 (3.11) | 13.5 (3.13) |
| Median [Min, Max] | 12.0 [10.0, 24.0] | 12.5 [10.0, 21.0] |
| Emotional Control | ||
| Mean (SD) | 15.0 (3.67) | 14.7 (3.04) |
| Median [Min, Max] | 14.0 [10.0, 24.0] | 14.0 [10.0, 21.0] |
| Working Memory | ||
| Mean (SD) | 22.7 (6.61) | 23.1 (4.80) |
| Median [Min, Max] | 20.5 [17.0, 48.0] | 22.0 [17.0, 37.0] |
| Plan/Organize | ||
| Mean (SD) | 14.3 (3.40) | 14.2 (3.03) |
| Median [Min, Max] | 13.5 [10.0, 23.0] | 14.0 [10.0, 20.0] |
t.test(BRIEF_early$GEC_RawScore~BRIEF_early$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early$GEC_RawScore by BRIEF_early$Language_Modality
## t = -0.40065, df = 58.665, p-value = 0.6901
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -10.065079 6.707219
## sample estimates:
## mean in group English mean in group ASL
## 87.78261 89.46154
t.test(BRIEF_early$Inhibit_RawScore~BRIEF_early$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early$Inhibit_RawScore by BRIEF_early$Language_Modality
## t = -0.81335, df = 51.079, p-value = 0.4198
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -4.164164 1.762826
## sample estimates:
## mean in group English mean in group ASL
## 22.76087 23.96154
t.test(BRIEF_early$Shift_RawScore~BRIEF_early$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early$Shift_RawScore by BRIEF_early$Language_Modality
## t = -0.57441, df = 51.763, p-value = 0.5682
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.976368 1.096769
## sample estimates:
## mean in group English mean in group ASL
## 13.02174 13.46154
t.test(BRIEF_early$Emotional.Control_RawScore~BRIEF_early$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early$Emotional.Control_RawScore by BRIEF_early$Language_Modality
## t = 0.30726, df = 60.411, p-value = 0.7597
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.363471 1.858454
## sample estimates:
## mean in group English mean in group ASL
## 14.97826 14.73077
t.test(BRIEF_early$Working.Memory_RawScore~BRIEF_early$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early$Working.Memory_RawScore by BRIEF_early$Language_Modality
## t = -0.29375, df = 65.462, p-value = 0.7699
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -3.103459 2.307473
## sample estimates:
## mean in group English mean in group ASL
## 22.71739 23.11538
t.test(BRIEF_early$Plan.Organize_RawScore~BRIEF_early$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early$Plan.Organize_RawScore by BRIEF_early$Language_Modality
## t = 0.14405, df = 57.092, p-value = 0.886
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.445442 1.669522
## sample estimates:
## mean in group English mean in group ASL
## 14.30435 14.19231
Findings: Early ASL group did not significantly differ from Typically Hearing group on any subscale ***
Approach: Linear regression models for the BRIEF-P raw scores on general composite score and each subscale predicted by age of auditory and age of language exposure, and other demographic characteristics (SES, sex, age)
Also testing whether other demographic variables or ways of characterizing group differences (Length of Auditory Exposure, Language Modality) are better predictive of BRIEF scores
Create dataframe for participants for whom we have Age of Auditory Exposure and Age of Language Exposure information
BRIEF <- dplyr::mutate(BRIEF, AoAE = ifelse(BRIEF$Age.of.Auditory.Exposure..mo.=="None", as.character(BRIEF$AgeMonths), as.character(BRIEF$Age.of.Auditory.Exposure..mo.)))
BRIEF$AoAE <- as.integer(BRIEF$AoAE)
BRIEF_AgeOf <- subset(BRIEF, BRIEF$AoAE!='' & BRIEF$Age.of.Language.Exposure..mo.!='')
table1::label(BRIEF_AgeOf$AgeMonths) <- "Age (Months)"
table1::label(BRIEF_AgeOf$SES..3.66.) <- "SES"
table1::label(BRIEF_AgeOf$Sex) <- "Sex"
table1::label(BRIEF_AgeOf$Race_recoded) <- "Race"
table1::label(BRIEF_AgeOf$Ethnicity_recoded) <- "Ethnicity"
table1::label(BRIEF_AgeOf$Age.of.Language.Exposure..mo.) <- "Age of Language Exposure (Months)"
table1::label(BRIEF_AgeOf$AoAE) <- "Age of Auditory Exposure (Months)"
table1::table1(~AgeMonths + SES..3.66. + Sex + Race_recoded + Ethnicity_recoded + Age.of.Language.Exposure..mo. + AoAE, data = BRIEF_AgeOf)
| Overall (N=109) |
|
|---|---|
| Age (Months) | |
| Mean (SD) | 59.2 (13.0) |
| Median [Min, Max] | 59.0 [37.0, 91.0] |
| SES | |
| Mean (SD) | 51.2 (13.5) |
| Median [Min, Max] | 54.5 [8.00, 66.0] |
| Sex | |
| Female | 58 (53.2%) |
| Male | 51 (46.8%) |
| Race | |
| Asian | 3 (2.8%) |
| Black or African American | 1 (0.9%) |
| White | 94 (86.2%) |
| More than one | 8 (7.3%) |
| Other/Missing | 3 (2.8%) |
| Ethnicity | |
| Hispanic | 6 (5.5%) |
| Non-Hispanic | 90 (82.6%) |
| Prefer not to answer | 1 (0.9%) |
| Missing | 12 (11.0%) |
| Age of Language Exposure (Months) | |
| Mean (SD) | 12.7 (18.8) |
| Median [Min, Max] | 0 [0, 76.0] |
| Age of Auditory Exposure (Months) | |
| Mean (SD) | 19.9 (25.2) |
| Median [Min, Max] | 4.00 [0, 91.0] |
table1::table1(~AgeMonths + SES..3.66. + Sex + Race_recoded + Ethnicity_recoded + Age.of.Language.Exposure..mo. + AoAE | LanguageGroup, data = BRIEF_AgeOf, overall=F)
| Typically Hearing (N=46) |
Early ASL (N=18) |
Later English (N=23) |
Later ASL (N=22) |
|
|---|---|---|---|---|
| Age (Months) | ||||
| Mean (SD) | 54.7 (10.9) | 59.2 (14.5) | 60.7 (10.9) | 67.3 (14.0) |
| Median [Min, Max] | 54.5 [37.0, 85.0] | 57.0 [41.0, 91.0] | 60.0 [37.0, 78.0] | 71.0 [37.0, 90.0] |
| SES | ||||
| Mean (SD) | 55.6 (8.83) | 49.0 (16.2) | 48.8 (14.1) | 46.0 (16.2) |
| Median [Min, Max] | 56.0 [21.5, 66.0] | 57.5 [19.0, 66.0] | 51.0 [8.00, 66.0] | 51.5 [14.0, 62.0] |
| Sex | ||||
| Female | 24 (52.2%) | 9 (50.0%) | 13 (56.5%) | 12 (54.5%) |
| Male | 22 (47.8%) | 9 (50.0%) | 10 (43.5%) | 10 (45.5%) |
| Race | ||||
| Asian | 0 (0%) | 0 (0%) | 0 (0%) | 3 (13.6%) |
| Black or African American | 0 (0%) | 0 (0%) | 1 (4.3%) | 0 (0%) |
| White | 43 (93.5%) | 17 (94.4%) | 18 (78.3%) | 16 (72.7%) |
| More than one | 3 (6.5%) | 0 (0%) | 3 (13.0%) | 2 (9.1%) |
| Other/Missing | 0 (0%) | 1 (5.6%) | 1 (4.3%) | 1 (4.5%) |
| Ethnicity | ||||
| Hispanic | 3 (6.5%) | 0 (0%) | 2 (8.7%) | 1 (4.5%) |
| Non-Hispanic | 42 (91.3%) | 13 (72.2%) | 17 (73.9%) | 18 (81.8%) |
| Prefer not to answer | 0 (0%) | 1 (5.6%) | 0 (0%) | 0 (0%) |
| Missing | 1 (2.2%) | 4 (22.2%) | 4 (17.4%) | 3 (13.6%) |
| Age of Language Exposure (Months) | ||||
| Mean (SD) | 0 (0) | 0 (0) | 21.2 (16.2) | 40.5 (12.5) |
| Median [Min, Max] | 0 [0, 0] | 0 [0, 0] | 16.0 [0.500, 58.0] | 36.5 [18.0, 76.0] |
| Age of Auditory Exposure (Months) | ||||
| Mean (SD) | 0 (0) | 49.0 (24.9) | 21.2 (16.3) | 36.3 (24.5) |
| Median [Min, Max] | 0 [0, 0] | 52.5 [1.00, 91.0] | 16.0 [0, 58.0] | 35.0 [3.00, 79.0] |
table1::label(BRIEF$Hearing_Level_3_Cat) <- "Pre-device hearing level (3 Category)"
table1::label(BRIEF$Hearing.Device) <- "Type of Hearing Device"
table1::label(BRIEF$AoAE) <- "Age of First Auditory Exposure (Months)"
table1::label(BRIEF$Age.of.Language.Exposure..mo.) <- "Age of First Language Exposure (Months)"
table1::table1(~Hearing_Level_3_Cat + Hearing.Device + AoAE + Age.of.Language.Exposure..mo.| LanguageGroup, data = BRIEF, overall=F)
| Typically Hearing (N=46) |
Early ASL (N=26) |
Later English (N=23) |
Later ASL (N=28) |
|
|---|---|---|---|---|
| Pre-device hearing level (3 Category) | ||||
| Typically Hearing | 46 (100%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Deaf | 0 (0%) | 15 (57.7%) | 13 (56.5%) | 21 (75.0%) |
| Hard of Hearing | 0 (0%) | 7 (26.9%) | 9 (39.1%) | 5 (17.9%) |
| Missing | 0 (0%) | 4 (15.4%) | 1 (4.3%) | 2 (7.1%) |
| Type of Hearing Device | ||||
| Hearing Aid | 0 (0%) | 11 (42.3%) | 11 (47.8%) | 12 (42.9%) |
| Cochlear Implant | 0 (0%) | 0 (0%) | 9 (39.1%) | 8 (28.6%) |
| Hearing Aid & Cochlear Implant | 0 (0%) | 0 (0%) | 2 (8.7%) | 1 (3.6%) |
| Other | 0 (0%) | 1 (3.8%) | 1 (4.3%) | 0 (0%) |
| None | 46 (100%) | 14 (53.8%) | 0 (0%) | 7 (25.0%) |
| Age of First Auditory Exposure (Months) | ||||
| Mean (SD) | 0 (0) | 49.0 (24.9) | 21.2 (16.3) | 38.6 (26.5) |
| Median [Min, Max] | 0 [0, 0] | 52.5 [1.00, 91.0] | 16.0 [0, 58.0] | 36.0 [3.00, 90.0] |
| Missing | 0 (0%) | 8 (30.8%) | 0 (0%) | 5 (17.9%) |
| Age of First Language Exposure (Months) | ||||
| Mean (SD) | 0 (0) | 0 (0) | 21.2 (16.2) | 42.0 (13.2) |
| Median [Min, Max] | 0 [0, 0] | 0 [0, 0] | 16.0 [0.500, 58.0] | 36.5 [18.0, 76.0] |
| Missing | 0 (0%) | 0 (0%) | 0 (0%) | 2 (7.1%) |
ggboxplot(BRIEF, x = "LanguageGroup", y = "AoAE",
color = "LanguageGroup",
order = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"),
ylab = "Age of Auditory Exposure (Months)", xlab = "Language Group")
## Warning: Removed 13 rows containing non-finite values (stat_boxplot).
ggplot(data=BRIEF, mapping=aes(x=AoAE))+ geom_histogram(binwidth=15) + facet_grid(~LanguageGroup)
## Warning: Removed 13 rows containing non-finite values (stat_bin).
leveneTest(AoAE~LanguageGroup, data=BRIEF)
## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 3 26.141 9.781e-13 ***
## 106
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
AoAE_LangGrp <- kruskal.test(AoAE~LanguageGroup, data=BRIEF)
AoAE_LangGrp
##
## Kruskal-Wallis rank sum test
##
## data: AoAE by LanguageGroup
## Kruskal-Wallis chi-squared = 88.967, df = 3, p-value < 2.2e-16
dunnTest(AoAE~LanguageGroup, data=BRIEF, method="hochberg")
## Warning: Some rows deleted from 'x' and 'g' because missing data.
## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Hochberg method.
## Comparison Z P.unadj P.adj
## 1 Early ASL - Later ASL 0.8923039 3.722301e-01 3.722301e-01
## 2 Early ASL - Later English 2.2593334 2.386265e-02 7.158796e-02
## 3 Later ASL - Later English 1.4588767 1.445991e-01 2.891981e-01
## 4 Early ASL - Typically Hearing 7.6412506 2.151219e-14 1.290732e-13
## 5 Later ASL - Typically Hearing 7.2191701 5.230578e-13 2.615289e-12
## 6 Later English - Typically Hearing 5.5346044 3.119315e-08 1.247726e-07
Findings:
ggboxplot(BRIEF, x = "LanguageGroup", y = "Age.of.Language.Exposure..mo.",
color = "LanguageGroup",
order = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"),
ylab = "Age of Language Exposure (Months)", xlab = "Language Group")
## Warning: Removed 2 rows containing non-finite values (stat_boxplot).
ggplot(data=BRIEF, mapping=aes(x=Age.of.Language.Exposure..mo.))+ geom_histogram(binwidth=10) + facet_grid(~LanguageGroup)
## Warning: Removed 2 rows containing non-finite values (stat_bin).
leveneTest(Age.of.Language.Exposure..mo.~LanguageGroup, data=BRIEF)
## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 3 29.968 1.9e-14 ***
## 117
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
AoLE_LangGrp <- kruskal.test(Age.of.Language.Exposure..mo.~LanguageGroup, data=BRIEF)
AoLE_LangGrp
##
## Kruskal-Wallis rank sum test
##
## data: Age.of.Language.Exposure..mo. by LanguageGroup
## Kruskal-Wallis chi-squared = 113.4, df = 3, p-value < 2.2e-16
dunnTest(Age.of.Language.Exposure..mo.~LanguageGroup, data=BRIEF, method="hochberg")
## Warning: Some rows deleted from 'x' and 'g' because missing data.
## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Hochberg method.
## Comparison Z P.unadj P.adj
## 1 Early ASL - Later ASL -7.894319 2.919049e-15 1.459524e-14
## 2 Early ASL - Later English -5.810672 6.222254e-09 1.866676e-08
## 3 Later ASL - Later English 1.838167 6.603775e-02 1.320755e-01
## 4 Early ASL - Typically Hearing 0.000000 1.000000e+00 1.000000e+00
## 5 Later ASL - Typically Hearing 8.923646 4.511844e-19 2.707106e-18
## 6 Later English - Typically Hearing 6.513165 7.358344e-11 2.943338e-10
Findings:
## Function below from: http://www.sthda.com/english/wiki/correlation-matrix-formatting-and-visualization
num_predictors <- BRIEF_AgeOf[, c("SES..3.66.", "AgeMonths", "Age.of.Language.Exposure..mo.", "AoAE")]
flattenCorrMatrix <- function(cormat, pmat) {
ut <- upper.tri(cormat)
data.frame(
row = rownames(cormat)[row(cormat)[ut]],
column = rownames(cormat)[col(cormat)[ut]],
cor =(cormat)[ut],
p = pmat[ut]
)
}
BRIEF_num_pred <- rcorr(as.matrix(num_predictors), type = c("pearson"))
flattenCorrMatrix(BRIEF_num_pred$r, BRIEF_num_pred$P)
## row column cor
## 1 SES..3.66. AgeMonths -0.002497801
## 2 SES..3.66. Age.of.Language.Exposure..mo. -0.244521622
## 3 AgeMonths Age.of.Language.Exposure..mo. 0.422545190
## 4 SES..3.66. AoAE -0.211797298
## 5 AgeMonths AoAE 0.399277901
## 6 Age.of.Language.Exposure..mo. AoAE 0.429319193
## p
## 1 9.794351e-01
## 2 1.039299e-02
## 3 4.719230e-06
## 4 2.704404e-02
## 5 1.701768e-05
## 6 3.189754e-06
Findings:
Step 1. Base Model (Demographic variables only):
overall_all_base <- lm(data=BRIEF_AgeOf, formula = GEC_RawScore~SES..3.66.+Sex+AgeMonths)
summary(overall_all_base)
##
## Call:
## lm(formula = GEC_RawScore ~ SES..3.66. + Sex + AgeMonths, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -32.330 -14.855 -0.360 8.628 60.531
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 107.1580 11.5903 9.245 3.02e-15 ***
## SES..3.66. -0.4345 0.1362 -3.191 0.00187 **
## SexMale -1.1793 3.6935 -0.319 0.75014
## AgeMonths 0.1167 0.1420 0.822 0.41320
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 18.94 on 105 degrees of freedom
## Multiple R-squared: 0.09454, Adjusted R-squared: 0.06867
## F-statistic: 3.654 on 3 and 105 DF, p-value: 0.01492
AIC(overall_all_base)
## [1] 956.5089
Step 2. Add Age of Language Exposure:
overall_all_AoLE <- lm(data=BRIEF_AgeOf, formula = GEC_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.)
summary(overall_all_AoLE)
##
## Call:
## lm(formula = GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -34.445 -13.455 -0.562 8.300 66.558
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 109.7397 11.2424 9.761 2.3e-16 ***
## SES..3.66. -0.3318 0.1364 -2.433 0.01667 *
## SexMale -1.8352 3.5785 -0.513 0.60916
## AgeMonths -0.0767 0.1529 -0.502 0.61687
## Age.of.Language.Exposure..mo. 0.3104 0.1077 2.883 0.00479 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 18.32 on 104 degrees of freedom
## Multiple R-squared: 0.1615, Adjusted R-squared: 0.1293
## F-statistic: 5.009 on 4 and 104 DF, p-value: 0.0009866
AIC(overall_all_AoLE)
## [1] 950.1289
Compare Model from Step 2 to model from Step 1:
anova(overall_all_base, overall_all_AoLE)
## Analysis of Variance Table
##
## Model 1: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 37686
## 2 104 34897 1 2788.7 8.3109 0.004789 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3a. Add Age of Auditory Exposure:
overall_all_AoLE_AoAE <- lm(data=BRIEF_AgeOf, formula = GEC_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+AoAE)
summary(overall_all_AoLE_AoAE)
##
## Call:
## lm(formula = GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -35.488 -13.012 -0.955 9.269 64.745
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 109.30858 11.24579 9.720 3.11e-16 ***
## SES..3.66. -0.35417 0.13801 -2.566 0.01172 *
## SexMale -1.70684 3.57928 -0.477 0.63447
## AgeMonths -0.02933 0.15945 -0.184 0.85440
## Age.of.Language.Exposure..mo. 0.34130 0.11166 3.057 0.00285 **
## AoAE -0.08470 0.08152 -1.039 0.30121
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 18.31 on 103 degrees of freedom
## Multiple R-squared: 0.1702, Adjusted R-squared: 0.13
## F-statistic: 4.226 on 5 and 103 DF, p-value: 0.001553
AIC(overall_all_AoLE_AoAE)
## [1] 950.9923
3b. Add Length of Auditory Experience:
overall_all_AoLE_LoAE <- lm(data=BRIEF_AgeOf, formula = GEC_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Length.of.Auditory.Exposure)
summary(overall_all_AoLE_LoAE)
##
## Call:
## lm(formula = GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Length.of.Auditory.Exposure, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -35.475 -13.038 -0.980 9.309 64.787
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 109.30137 11.24709 9.718 3.13e-16 ***
## SES..3.66. -0.35384 0.13800 -2.564 0.01179 *
## SexMale -1.70954 3.57952 -0.478 0.63395
## AgeMonths -0.11363 0.15695 -0.724 0.47071
## Age.of.Language.Exposure..mo. 0.34085 0.11162 3.054 0.00288 **
## Length.of.Auditory.Exposure 0.08405 0.08153 1.031 0.30502
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 18.31 on 103 degrees of freedom
## Multiple R-squared: 0.1701, Adjusted R-squared: 0.1298
## F-statistic: 4.222 on 5 and 103 DF, p-value: 0.001565
AIC(overall_all_AoLE_LoAE)
## [1] 951.0101
3c. Add Modality:
overall_all_AoLE_Modality <- lm(data=BRIEF_AgeOf, formula = GEC_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Language_Modality)
summary(overall_all_AoLE_Modality)
##
## Call:
## lm(formula = GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Language_Modality, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -34.164 -12.261 -1.154 8.590 65.839
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 110.23865 11.32017 9.738 2.83e-16 ***
## SES..3.66. -0.34263 0.13834 -2.477 0.01489 *
## SexMale -1.82344 3.59098 -0.508 0.61269
## AgeMonths -0.06616 0.15465 -0.428 0.66967
## Age.of.Language.Exposure..mo. 0.32718 0.11254 2.907 0.00447 **
## Language_ModalityASL -2.15194 4.04013 -0.533 0.59543
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 18.38 on 103 degrees of freedom
## Multiple R-squared: 0.1638, Adjusted R-squared: 0.1233
## F-statistic: 4.037 on 5 and 103 DF, p-value: 0.002191
AIC(overall_all_AoLE_Modality)
## [1] 951.8291
3d.Add Hearing Status (3-category):
overall_all_AoLE_HearStat <- lm(data=BRIEF_AgeOf, formula = GEC_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+ Hearing_Level_3_Cat)
summary(overall_all_AoLE_HearStat)
##
## Call:
## lm(formula = GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Hearing_Level_3_Cat, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -33.895 -12.842 -0.688 8.396 65.922
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 112.3796 12.2077 9.206 7.57e-15 ***
## SES..3.66. -0.3348 0.1430 -2.341 0.0213 *
## SexMale -0.6712 3.7204 -0.180 0.8572
## AgeMonths -0.1033 0.1686 -0.613 0.5415
## Age.of.Language.Exposure..mo. 0.3130 0.1316 2.378 0.0194 *
## Hearing_Level_3_CatDeaf -2.3984 4.9761 -0.482 0.6309
## Hearing_Level_3_CatHard of Hearing -2.2487 6.3660 -0.353 0.7247
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 18.48 on 96 degrees of freedom
## (6 observations deleted due to missingness)
## Multiple R-squared: 0.1434, Adjusted R-squared: 0.08983
## F-statistic: 2.678 on 6 and 96 DF, p-value: 0.01905
AIC(overall_all_AoLE_HearStat)
## [1] 901.8779
Compare Models from Step 3a-c to model from Step 2:
anova(overall_all_AoLE, overall_all_AoLE_AoAE)
## Analysis of Variance Table
##
## Model 1: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 34897
## 2 103 34535 1 361.99 1.0796 0.3012
anova(overall_all_AoLE, overall_all_AoLE_Modality)
## Analysis of Variance Table
##
## Model 1: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Language_Modality
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 34897
## 2 103 34801 1 95.857 0.2837 0.5954
anova(overall_all_AoLE, overall_all_AoLE_LoAE)
## Analysis of Variance Table
##
## Model 1: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Length.of.Auditory.Exposure
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 34897
## 2 103 34541 1 356.36 1.0627 0.305
Findings: Age of auditory exposure, length of auditory exposure, and language modality are not significant predictors of GEC raw scores, and their addition does not improve model fit. We cannot do a direct comparison of model in step 3d to the model in step 2 because we are missing hearing status information for 12 participants (mostly Early and Later ASL). However, even in the model in step 3d, Hearing Status is not a significant predictor of BRIEF scores, and Age of Language Exposure is.
Checking assumptions for best-fit GEC model
ols_plot_resid_qq(overall_all_AoLE)
ols_test_normality(overall_all_AoLE)
## -----------------------------------------------
## Test Statistic pvalue
## -----------------------------------------------
## Shapiro-Wilk 0.9674 0.0089
## Kolmogorov-Smirnov 0.0841 0.4242
## Cramer-von Mises 8.634 0.0000
## Anderson-Darling 0.8395 0.0296
## -----------------------------------------------
ols_plot_resid_hist(overall_all_AoLE)
ols_test_correlation(overall_all_AoLE)
## [1] 0.9826969
ols_plot_resid_fit(overall_all_AoLE)
Step 1. Base Model:
inhibition_all_base <- lm(data=BRIEF_AgeOf, formula = Inhibit_RawScore~SES..3.66.+Sex+AgeMonths)
summary(inhibition_all_base)
##
## Call:
## lm(formula = Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.4961 -4.5032 0.0108 4.0216 21.5107
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.07731 3.65405 6.863 4.86e-10 ***
## SES..3.66. -0.10990 0.04292 -2.560 0.0119 *
## SexMale 0.97828 1.16445 0.840 0.4028
## AgeMonths 0.06696 0.04478 1.495 0.1378
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.973 on 105 degrees of freedom
## Multiple R-squared: 0.08539, Adjusted R-squared: 0.05926
## F-statistic: 3.268 on 3 and 105 DF, p-value: 0.02424
AIC(inhibition_all_base)
## [1] 704.8643
Step 2. Add Age of Language Exposure:
inhibition_all_AoLE <- lm(data=BRIEF_AgeOf, formula = Inhibit_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.)
summary(inhibition_all_AoLE)
##
## Call:
## lm(formula = Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.9261 -4.2881 -0.1993 4.2050 22.7360
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.60218 3.62617 7.060 1.93e-10 ***
## SES..3.66. -0.08903 0.04399 -2.024 0.0455 *
## SexMale 0.84495 1.15423 0.732 0.4658
## AgeMonths 0.02764 0.04930 0.561 0.5762
## Age.of.Language.Exposure..mo. 0.06310 0.03473 1.817 0.0721 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.908 on 104 degrees of freedom
## Multiple R-squared: 0.1135, Adjusted R-squared: 0.07944
## F-statistic: 3.33 on 4 and 104 DF, p-value: 0.01311
AIC(inhibition_all_AoLE)
## [1] 703.4576
Compare Model with Age of Language Exposure to base demographic model:
anova(inhibition_all_base, inhibition_all_AoLE)
## Analysis of Variance Table
##
## Model 1: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 3745.7
## 2 104 3630.5 1 115.26 3.3018 0.07208 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3a. Add Age of Auditory Experience:
inhibition_all_AoLE_AoAE <- lm(data=BRIEF_AgeOf, formula = Inhibit_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+AoAE)
summary(inhibition_all_AoLE_AoAE)
##
## Call:
## lm(formula = Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.3417 -4.3176 -0.1074 4.1316 22.2424
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.48478 3.63271 7.015 2.49e-10 ***
## SES..3.66. -0.09511 0.04458 -2.133 0.0353 *
## SexMale 0.87988 1.15621 0.761 0.4484
## AgeMonths 0.04054 0.05151 0.787 0.4330
## Age.of.Language.Exposure..mo. 0.07152 0.03607 1.983 0.0501 .
## AoAE -0.02306 0.02633 -0.876 0.3832
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.915 on 103 degrees of freedom
## Multiple R-squared: 0.1201, Adjusted R-squared: 0.07737
## F-statistic: 2.811 on 5 and 103 DF, p-value: 0.02018
AIC(inhibition_all_AoLE_AoAE)
## [1] 704.6489
3b. Add Length of Auditory Exposure:
inhibition_all_AoLE_LoAE <- lm(data=BRIEF_AgeOf, formula = Inhibit_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Length.of.Auditory.Exposure)
summary(inhibition_all_AoLE_LoAE)
##
## Call:
## lm(formula = Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Length.of.Auditory.Exposure,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.3466 -4.3172 -0.0975 4.1237 22.2565
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.48347 3.63318 7.014 2.5e-10 ***
## SES..3.66. -0.09498 0.04458 -2.131 0.0355 *
## SexMale 0.87896 1.15630 0.760 0.4489
## AgeMonths 0.01764 0.05070 0.348 0.7285
## Age.of.Language.Exposure..mo. 0.07135 0.03606 1.979 0.0505 .
## Length.of.Auditory.Exposure 0.02276 0.02634 0.864 0.3895
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.916 on 103 degrees of freedom
## Multiple R-squared: 0.1199, Adjusted R-squared: 0.07719
## F-statistic: 2.807 on 5 and 103 DF, p-value: 0.02035
AIC(inhibition_all_AoLE_LoAE)
## [1] 704.6701
3c. Add Modality:
inhibition_all_AoLE_Modality <- lm(data=BRIEF_AgeOf, formula = Inhibit_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+ Language_Modality)
summary(inhibition_all_AoLE_Modality)
##
## Call:
## lm(formula = Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Language_Modality, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.9319 -4.2811 -0.2139 4.2162 22.7509
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.59185 3.65624 6.999 2.69e-10 ***
## SES..3.66. -0.08880 0.04468 -1.987 0.0495 *
## SexMale 0.84470 1.15983 0.728 0.4681
## AgeMonths 0.02743 0.04995 0.549 0.5841
## Age.of.Language.Exposure..mo. 0.06276 0.03635 1.726 0.0873 .
## Language_ModalityASL 0.04455 1.30490 0.034 0.9728
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.937 on 103 degrees of freedom
## Multiple R-squared: 0.1135, Adjusted R-squared: 0.07051
## F-statistic: 2.639 on 5 and 103 DF, p-value: 0.02752
AIC(inhibition_all_AoLE_Modality)
## [1] 705.4563
3d. Add Hearing Status (3-category):
inhibition_all_AoLE_HearStat <- lm(data=BRIEF_AgeOf, formula = Inhibit_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+ Hearing_Level_3_Cat)
summary(inhibition_all_AoLE_HearStat)
##
## Call:
## lm(formula = Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Hearing_Level_3_Cat, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.6102 -4.5798 -0.0519 3.7536 22.9804
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 26.361868 3.902861 6.754 1.1e-09 ***
## SES..3.66. -0.082412 0.045729 -1.802 0.0747 .
## SexMale 1.051729 1.189446 0.884 0.3788
## AgeMonths 0.008764 0.053899 0.163 0.8712
## Age.of.Language.Exposure..mo. 0.049904 0.042073 1.186 0.2385
## Hearing_Level_3_CatDeaf 0.620825 1.590893 0.390 0.6972
## Hearing_Level_3_CatHard of Hearing -1.545955 2.035248 -0.760 0.4494
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.908 on 96 degrees of freedom
## (6 observations deleted due to missingness)
## Multiple R-squared: 0.1096, Adjusted R-squared: 0.0539
## F-statistic: 1.969 on 6 and 96 DF, p-value: 0.07767
AIC(inhibition_all_AoLE_HearStat)
## [1] 666.9651
Compare Models from Step 3a-c to model from Step 2:
anova(inhibition_all_AoLE, inhibition_all_AoLE_AoAE)
## Analysis of Variance Table
##
## Model 1: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 3630.5
## 2 103 3603.6 1 26.837 0.7671 0.3832
anova(inhibition_all_AoLE, inhibition_all_AoLE_Modality)
## Analysis of Variance Table
##
## Model 1: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Language_Modality
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 3630.5
## 2 103 3630.4 1 0.041089 0.0012 0.9728
anova(inhibition_all_AoLE, inhibition_all_AoLE_LoAE)
## Analysis of Variance Table
##
## Model 1: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Length.of.Auditory.Exposure
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 3630.5
## 2 103 3604.3 1 26.134 0.7468 0.3895
Checking assumptions for best-fit Inhibition model
ols_plot_resid_qq(inhibition_all_AoLE)
ols_test_normality(inhibition_all_AoLE)
## -----------------------------------------------
## Test Statistic pvalue
## -----------------------------------------------
## Shapiro-Wilk 0.9723 0.0226
## Kolmogorov-Smirnov 0.0732 0.6029
## Cramer-von Mises 8.308 0.0000
## Anderson-Darling 0.5534 0.1502
## -----------------------------------------------
ols_plot_resid_hist(inhibition_all_AoLE)
ols_test_correlation(inhibition_all_AoLE)
## [1] 0.9845734
ols_plot_resid_fit(inhibition_all_AoLE)
Step 1. Base Model:
shift_all_base <- lm(data=BRIEF_AgeOf, formula = Shift_RawScore~SES..3.66.+Sex+AgeMonths)
summary(shift_all_base)
##
## Call:
## lm(formula = Shift_RawScore ~ SES..3.66. + Sex + AgeMonths, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.936 -2.682 -0.776 1.774 14.985
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 17.93727 2.20373 8.140 8.65e-13 ***
## SES..3.66. -0.08380 0.02589 -3.237 0.00162 **
## SexMale -0.79353 0.70227 -1.130 0.26107
## AgeMonths 0.01077 0.02701 0.399 0.69088
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.602 on 105 degrees of freedom
## Multiple R-squared: 0.0979, Adjusted R-squared: 0.07213
## F-statistic: 3.799 on 3 and 105 DF, p-value: 0.01245
AIC(shift_all_base)
## [1] 594.6247
Step 2. Add Age of Language Exposure:
shift_all_AoLE <- lm(data=BRIEF_AgeOf, formula = Shift_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.)
summary(shift_all_AoLE)
##
## Call:
## lm(formula = Shift_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.5170 -2.2586 -0.6156 1.4734 14.0409
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18.43971 2.13351 8.643 7.11e-14 ***
## SES..3.66. -0.06382 0.02588 -2.466 0.01530 *
## SexMale -0.92116 0.67911 -1.356 0.17790
## AgeMonths -0.02687 0.02901 -0.926 0.35646
## Age.of.Language.Exposure..mo. 0.06041 0.02043 2.956 0.00385 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.476 on 104 degrees of freedom
## Multiple R-squared: 0.1678, Adjusted R-squared: 0.1358
## F-statistic: 5.244 on 4 and 104 DF, p-value: 0.0006899
AIC(shift_all_AoLE)
## [1] 587.829
Compare Model with Age of Language Exposure to base demographic model:
anova(shift_all_base, shift_all_AoLE)
## Analysis of Variance Table
##
## Model 1: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 1362.4
## 2 104 1256.8 1 105.62 8.7401 0.003852 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3a. Add Age of Auditory Experience:
shift_all_AoLE_AoAE <- lm(data=BRIEF_AgeOf, formula = Shift_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+AoAE)
summary(shift_all_AoLE_AoAE)
##
## Call:
## lm(formula = Shift_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.4369 -2.2772 -0.5143 1.4576 14.0693
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18.425943 2.144990 8.590 9.89e-14 ***
## SES..3.66. -0.064533 0.026323 -2.452 0.0159 *
## SexMale -0.917061 0.682702 -1.343 0.1821
## AgeMonths -0.025355 0.030413 -0.834 0.4064
## Age.of.Language.Exposure..mo. 0.061394 0.021298 2.883 0.0048 **
## AoAE -0.002704 0.015549 -0.174 0.8623
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.493 on 103 degrees of freedom
## Multiple R-squared: 0.1681, Adjusted R-squared: 0.1277
## F-statistic: 4.162 on 5 and 103 DF, p-value: 0.001745
AIC(shift_all_AoLE_AoAE)
## [1] 589.797
3b. Add Length of Auditory Exposure:
shift_all_AoLE_LoAE <- lm(data=BRIEF_AgeOf, formula = Shift_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Length.of.Auditory.Exposure)
summary(shift_all_AoLE_LoAE)
##
## Call:
## lm(formula = Shift_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Length.of.Auditory.Exposure,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.4408 -2.2740 -0.5135 1.4587 14.0684
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18.426317 2.145094 8.590 9.91e-14 ***
## SES..3.66. -0.064492 0.026319 -2.450 0.01596 *
## SexMale -0.917321 0.682701 -1.344 0.18201
## AgeMonths -0.027996 0.029934 -0.935 0.35185
## Age.of.Language.Exposure..mo. 0.061337 0.021288 2.881 0.00482 **
## Length.of.Auditory.Exposure 0.002567 0.015550 0.165 0.86918
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.493 on 103 degrees of freedom
## Multiple R-squared: 0.1681, Adjusted R-squared: 0.1277
## F-statistic: 4.161 on 5 and 103 DF, p-value: 0.001748
AIC(shift_all_AoLE_LoAE)
## [1] 589.8002
3c. Add Modality:
shift_all_AoLE_Modality <- lm(data=BRIEF_AgeOf, formula = Shift_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Language_Modality)
summary(shift_all_AoLE_Modality)
##
## Call:
## lm(formula = Shift_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Language_Modality, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.5425 -2.2325 -0.6918 1.4192 14.0009
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18.41341 2.15099 8.560 1.15e-13 ***
## SES..3.66. -0.06325 0.02629 -2.406 0.0179 *
## SexMale -0.92178 0.68234 -1.351 0.1797
## AgeMonths -0.02742 0.02939 -0.933 0.3529
## Age.of.Language.Exposure..mo. 0.05952 0.02138 2.783 0.0064 **
## Language_ModalityASL 0.11343 0.76768 0.148 0.8828
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.493 on 103 degrees of freedom
## Multiple R-squared: 0.168, Adjusted R-squared: 0.1276
## F-statistic: 4.16 on 5 and 103 DF, p-value: 0.001752
AIC(shift_all_AoLE_Modality)
## [1] 589.8059
3d. Add Hearing Status (3-category):
shift_all_AoLE_HearStat <- lm(data=BRIEF_AgeOf, formula = Shift_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+ Hearing_Level_3_Cat)
summary(shift_all_AoLE_HearStat)
##
## Call:
## lm(formula = Shift_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Hearing_Level_3_Cat, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.9789 -2.2795 -0.4574 1.7698 12.7067
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18.39764 2.30188 7.992 2.95e-12 ***
## SES..3.66. -0.07108 0.02697 -2.635 0.0098 **
## SexMale -0.69119 0.70153 -0.985 0.3270
## AgeMonths -0.01995 0.03179 -0.628 0.5318
## Age.of.Language.Exposure..mo. 0.04517 0.02481 1.820 0.0718 .
## Hearing_Level_3_CatDeaf -0.19032 0.93830 -0.203 0.8397
## Hearing_Level_3_CatHard of Hearing 1.76805 1.20037 1.473 0.1440
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.484 on 96 degrees of freedom
## (6 observations deleted due to missingness)
## Multiple R-squared: 0.1804, Adjusted R-squared: 0.1291
## F-statistic: 3.521 on 6 and 96 DF, p-value: 0.003423
AIC(shift_all_AoLE_HearStat)
## [1] 558.2004
Compare Models from Step 3a-c to model from Step 2:
anova(shift_all_AoLE, shift_all_AoLE_AoAE)
## Analysis of Variance Table
##
## Model 1: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1256.8
## 2 103 1256.4 1 0.369 0.0303 0.8623
anova(shift_all_AoLE, shift_all_AoLE_Modality)
## Analysis of Variance Table
##
## Model 1: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Language_Modality
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1256.8
## 2 103 1256.5 1 0.26633 0.0218 0.8828
anova(shift_all_AoLE, shift_all_AoLE_LoAE)
## Analysis of Variance Table
##
## Model 1: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Length.of.Auditory.Exposure
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1256.8
## 2 103 1256.4 1 0.33256 0.0273 0.8692
Checking assumptions for best-fit Shift model
ols_plot_resid_qq(shift_all_AoLE)
ols_test_normality(shift_all_AoLE)
## -----------------------------------------------
## Test Statistic pvalue
## -----------------------------------------------
## Shapiro-Wilk 0.9237 0.0000
## Kolmogorov-Smirnov 0.0984 0.2423
## Cramer-von Mises 9.0788 0.0000
## Anderson-Darling 1.7709 1e-04
## -----------------------------------------------
ols_plot_resid_hist(shift_all_AoLE)
ols_test_correlation(shift_all_AoLE)
## [1] 0.9589466
ols_plot_resid_fit(shift_all_AoLE)
Step 1. Base Model:
emotctrl_all_base <- lm(data=BRIEF_AgeOf, formula = Emotional.Control_RawScore~SES..3.66.+Sex+AgeMonths)
summary(emotctrl_all_base)
##
## Call:
## lm(formula = Emotional.Control_RawScore ~ SES..3.66. + Sex +
## AgeMonths, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.4561 -3.2273 -0.6821 2.4962 13.8386
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.049467 2.498494 8.025 1.55e-12 ***
## SES..3.66. -0.091365 0.029350 -3.113 0.00239 **
## SexMale -0.705779 0.796204 -0.886 0.37741
## AgeMonths 0.001859 0.030620 0.061 0.95170
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.084 on 105 degrees of freedom
## Multiple R-squared: 0.08711, Adjusted R-squared: 0.06103
## F-statistic: 3.34 on 3 and 105 DF, p-value: 0.02214
Step 2. Add Age of Language Exposure:
emotctrl_all_AoLE <- lm(data=BRIEF_AgeOf, formula = Emotional.Control_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.)
summary(emotctrl_all_AoLE)
##
## Call:
## lm(formula = Emotional.Control_RawScore ~ SES..3.66. + Sex +
## AgeMonths + Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.1523 -3.0899 -0.5234 2.6179 13.4122
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.27630 2.50295 8.101 1.11e-12 ***
## SES..3.66. -0.08234 0.03036 -2.712 0.00783 **
## SexMale -0.76340 0.79670 -0.958 0.34018
## AgeMonths -0.01513 0.03403 -0.445 0.65747
## Age.of.Language.Exposure..mo. 0.02727 0.02397 1.138 0.25786
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.078 on 104 degrees of freedom
## Multiple R-squared: 0.09833, Adjusted R-squared: 0.06365
## F-statistic: 2.836 on 4 and 104 DF, p-value: 0.0281
Compare Model with Age of Language Exposure to base demographic model:
anova(emotctrl_all_base, emotctrl_all_AoLE)
## Analysis of Variance Table
##
## Model 1: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 1751.2
## 2 104 1729.7 1 21.528 1.2944 0.2579
3a. Add Age of Auditory Experience:
emotctrl_all_AoLE_AoAE <- lm(data=BRIEF_AgeOf, formula = Emotional.Control_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+AoAE)
summary(emotctrl_all_AoLE_AoAE)
##
## Call:
## lm(formula = Emotional.Control_RawScore ~ SES..3.66. + Sex +
## AgeMonths + Age.of.Language.Exposure..mo. + AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.418 -3.282 -0.533 2.551 13.559
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.204803 2.509534 8.051 1.5e-12 ***
## SES..3.66. -0.086045 0.030797 -2.794 0.00621 **
## SexMale -0.742122 0.798728 -0.929 0.35499
## AgeMonths -0.007278 0.035581 -0.205 0.83833
## Age.of.Language.Exposure..mo. 0.032397 0.024918 1.300 0.19645
## AoAE -0.014046 0.018191 -0.772 0.44179
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.086 on 103 degrees of freedom
## Multiple R-squared: 0.1035, Adjusted R-squared: 0.06
## F-statistic: 2.379 on 5 and 103 DF, p-value: 0.0437
3b. Add Length of Auditory Experience:
emotctrl_all_AoLE_LoAE <- lm(data=BRIEF_AgeOf, formula = Emotional.Control_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Length.of.Auditory.Exposure)
summary(emotctrl_all_AoLE_LoAE)
##
## Call:
## lm(formula = Emotional.Control_RawScore ~ SES..3.66. + Sex +
## AgeMonths + Age.of.Language.Exposure..mo. + Length.of.Auditory.Exposure,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.421 -3.278 -0.527 2.557 13.562
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.20351 2.50971 8.050 1.51e-12 ***
## SES..3.66. -0.08599 0.03079 -2.793 0.00623 **
## SexMale -0.74254 0.79875 -0.930 0.35473
## AgeMonths -0.02127 0.03502 -0.607 0.54506
## Age.of.Language.Exposure..mo. 0.03233 0.02491 1.298 0.19719
## Length.of.Auditory.Exposure 0.01396 0.01819 0.767 0.44474
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.086 on 103 degrees of freedom
## Multiple R-squared: 0.1035, Adjusted R-squared: 0.05994
## F-statistic: 2.377 on 5 and 103 DF, p-value: 0.04383
3c. Add Modality:
emotctrl_all_AoLE_Modality <- lm(data=BRIEF_AgeOf, formula = Emotional.Control_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Language_Modality)
summary(emotctrl_all_AoLE_Modality)
##
## Call:
## lm(formula = Emotional.Control_RawScore ~ SES..3.66. + Sex +
## AgeMonths + Age.of.Language.Exposure..mo. + Language_Modality,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.2712 -3.1432 -0.4244 2.5967 13.4687
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.31350 2.52334 8.050 1.51e-12 ***
## SES..3.66. -0.08315 0.03084 -2.696 0.00819 **
## SexMale -0.76253 0.80045 -0.953 0.34301
## AgeMonths -0.01435 0.03447 -0.416 0.67813
## Age.of.Language.Exposure..mo. 0.02852 0.02509 1.137 0.25818
## Language_ModalityASL -0.16042 0.90057 -0.178 0.85897
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.097 on 103 degrees of freedom
## Multiple R-squared: 0.09861, Adjusted R-squared: 0.05485
## F-statistic: 2.254 on 5 and 103 DF, p-value: 0.0545
3d. Add Hearing Status (3-category):
emotctrl_all_AoLE_HearStat <- lm(data=BRIEF_AgeOf, formula = Emotional.Control_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+ Hearing_Level_3_Cat)
summary(emotctrl_all_AoLE_HearStat)
##
## Call:
## lm(formula = Emotional.Control_RawScore ~ SES..3.66. + Sex +
## AgeMonths + Age.of.Language.Exposure..mo. + Hearing_Level_3_Cat,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -7.7400 -2.9737 -0.7746 2.3407 13.1880
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.89805 2.71573 7.695 1.25e-11 ***
## SES..3.66. -0.08644 0.03182 -2.716 0.00783 **
## SexMale -0.56179 0.82765 -0.679 0.49892
## AgeMonths -0.01541 0.03750 -0.411 0.68210
## Age.of.Language.Exposure..mo. 0.03813 0.02928 1.302 0.19593
## Hearing_Level_3_CatDeaf -1.24135 1.10699 -1.121 0.26493
## Hearing_Level_3_CatHard of Hearing -0.59943 1.41619 -0.423 0.67304
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.111 on 96 degrees of freedom
## (6 observations deleted due to missingness)
## Multiple R-squared: 0.09777, Adjusted R-squared: 0.04138
## F-statistic: 1.734 on 6 and 96 DF, p-value: 0.1213
Compare Models from Step 3a-c to model from Step 2:
anova(emotctrl_all_AoLE, emotctrl_all_AoLE_AoAE)
## Analysis of Variance Table
##
## Model 1: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1729.7
## 2 103 1719.8 1 9.9548 0.5962 0.4418
anova(emotctrl_all_AoLE, emotctrl_all_AoLE_Modality)
## Analysis of Variance Table
##
## Model 1: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Language_Modality
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1729.7
## 2 103 1729.2 1 0.53271 0.0317 0.859
anova(emotctrl_all_AoLE, emotctrl_all_AoLE_LoAE)
## Analysis of Variance Table
##
## Model 1: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Length.of.Auditory.Exposure
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1729.7
## 2 103 1719.9 1 9.8272 0.5885 0.4447
Checking assumptions for best-fit Emotional Control model
ols_plot_resid_qq(emotctrl_all_AoLE)
ols_test_normality(emotctrl_all_AoLE)
## -----------------------------------------------
## Test Statistic pvalue
## -----------------------------------------------
## Shapiro-Wilk 0.9536 8e-04
## Kolmogorov-Smirnov 0.1098 0.1447
## Cramer-von Mises 10.0351 0.0000
## Anderson-Darling 1.4899 7e-04
## -----------------------------------------------
ols_plot_resid_hist(emotctrl_all_AoLE)
ols_test_correlation(emotctrl_all_AoLE)
## [1] 0.9762239
ols_plot_resid_fit(emotctrl_all_AoLE)
Step 1. Base Model:
wm_all_base <- lm(data=BRIEF_AgeOf, formula = Working.Memory_RawScore~SES..3.66.+Sex+AgeMonths)
summary(wm_all_base)
##
## Call:
## lm(formula = Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.748 -4.607 -1.414 3.432 23.472
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.20501 3.70291 6.807 6.38e-10 ***
## SES..3.66. -0.07306 0.04350 -1.680 0.096 .
## SexMale -0.43333 1.18002 -0.367 0.714
## AgeMonths 0.04043 0.04538 0.891 0.375
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 6.053 on 105 degrees of freedom
## Multiple R-squared: 0.0346, Adjusted R-squared: 0.007015
## F-statistic: 1.254 on 3 and 105 DF, p-value: 0.294
AIC(wm_all_base)
## [1] 707.7603
Step 2. Add Age of Language Exposure:
wm_all_AoLE <- lm(data=BRIEF_AgeOf, formula = Working.Memory_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.)
summary(wm_all_AoLE)
##
## Call:
## lm(formula = Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -9.445 -4.061 -1.545 3.381 25.457
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 26.05515 3.58282 7.272 6.86e-11 ***
## SES..3.66. -0.03925 0.04346 -0.903 0.3686
## SexMale -0.64928 1.14043 -0.569 0.5704
## AgeMonths -0.02326 0.04871 -0.477 0.6341
## Age.of.Language.Exposure..mo. 0.10221 0.03431 2.979 0.0036 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.838 on 104 degrees of freedom
## Multiple R-squared: 0.1105, Adjusted R-squared: 0.07628
## F-statistic: 3.23 on 4 and 104 DF, p-value: 0.01531
AIC(wm_all_AoLE)
## [1] 700.8361
Compare Model with Age of Language Exposure to base demographic model:
anova(wm_all_base, wm_all_AoLE)
## Analysis of Variance Table
##
## Model 1: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 3846.6
## 2 104 3544.2 1 302.39 8.8732 0.003602 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3a. Add Age of Auditory Experience:
wm_all_AoLE_AoAE <- lm(data=BRIEF_AgeOf, formula = Working.Memory_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+AoAE)
summary(wm_all_AoLE_AoAE)
##
## Call:
## lm(formula = Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -9.313 -4.115 -1.080 3.208 24.929
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.929624 3.586996 7.229 8.82e-11 ***
## SES..3.66. -0.045745 0.044020 -1.039 0.30115
## SexMale -0.611926 1.141660 -0.536 0.59312
## AgeMonths -0.009465 0.050858 -0.186 0.85272
## Age.of.Language.Exposure..mo. 0.111208 0.035616 3.122 0.00233 **
## AoAE -0.024660 0.026002 -0.948 0.34514
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.841 on 103 degrees of freedom
## Multiple R-squared: 0.1182, Adjusted R-squared: 0.07538
## F-statistic: 2.761 on 5 and 103 DF, p-value: 0.02209
AIC(wm_all_AoLE_AoAE)
## [1] 701.8883
3b. Add Length of Auditory Experience:
wm_all_AoLE_LoAE <- lm(data=BRIEF_AgeOf, formula = Working.Memory_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Length.of.Auditory.Exposure)
summary(wm_all_AoLE_LoAE)
##
## Call:
## lm(formula = Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Length.of.Auditory.Exposure,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -9.311 -4.115 -1.085 3.214 24.940
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.92725 3.58730 7.228 8.87e-11 ***
## SES..3.66. -0.04566 0.04401 -1.037 0.30197
## SexMale -0.61263 1.14170 -0.537 0.59270
## AgeMonths -0.03403 0.05006 -0.680 0.49816
## Age.of.Language.Exposure..mo. 0.11109 0.03560 3.121 0.00234 **
## Length.of.Auditory.Exposure 0.02452 0.02600 0.943 0.34791
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.841 on 103 degrees of freedom
## Multiple R-squared: 0.1181, Adjusted R-squared: 0.07529
## F-statistic: 2.759 on 5 and 103 DF, p-value: 0.02219
AIC(wm_all_AoLE_LoAE)
## [1] 701.8992
3c. Add Modality:
wm_all_AoLE_Modality <- lm(data=BRIEF_AgeOf, formula = Working.Memory_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Language_Modality)
summary(wm_all_AoLE_Modality)
##
## Call:
## lm(formula = Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Language_Modality, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -9.328 -4.082 -1.427 3.255 25.158
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 26.26220 3.60414 7.287 6.64e-11 ***
## SES..3.66. -0.04372 0.04404 -0.993 0.32322
## SexMale -0.64442 1.14330 -0.564 0.57422
## AgeMonths -0.01888 0.04924 -0.384 0.70213
## Age.of.Language.Exposure..mo. 0.10918 0.03583 3.047 0.00294 **
## Language_ModalityASL -0.89303 1.28630 -0.694 0.48908
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.852 on 103 degrees of freedom
## Multiple R-squared: 0.1146, Adjusted R-squared: 0.07165
## F-statistic: 2.667 on 5 and 103 DF, p-value: 0.02615
AIC(wm_all_AoLE_Modality)
## [1] 702.3272
3d. Add Hearing Status (3-category):
wm_all_AoLE_HearStat <- lm(data=BRIEF_AgeOf, formula = Working.Memory_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+ Hearing_Level_3_Cat)
summary(wm_all_AoLE_HearStat)
##
## Call:
## lm(formula = Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Hearing_Level_3_Cat, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -9.4405 -4.3817 -0.8562 2.9612 25.3821
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 26.69333 3.89045 6.861 6.67e-10 ***
## SES..3.66. -0.03277 0.04558 -0.719 0.47391
## SexMale -0.41046 1.18566 -0.346 0.72996
## AgeMonths -0.03575 0.05373 -0.665 0.50742
## Age.of.Language.Exposure..mo. 0.11179 0.04194 2.665 0.00902 **
## Hearing_Level_3_CatDeaf -0.44075 1.58583 -0.278 0.78167
## Hearing_Level_3_CatHard of Hearing -1.82749 2.02878 -0.901 0.36996
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.889 on 96 degrees of freedom
## (6 observations deleted due to missingness)
## Multiple R-squared: 0.1026, Adjusted R-squared: 0.04653
## F-statistic: 1.83 on 6 and 96 DF, p-value: 0.1013
AIC(wm_all_AoLE_HearStat)
## [1] 666.309
Compare Models from Step 3a-c to model from Step 2:
anova(wm_all_AoLE, wm_all_AoLE_AoAE)
## Analysis of Variance Table
##
## Model 1: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 3544.2
## 2 103 3513.5 1 30.683 0.8995 0.3451
anova(wm_all_AoLE, wm_all_AoLE_Modality)
## Analysis of Variance Table
##
## Model 1: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Language_Modality
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 3544.2
## 2 103 3527.7 1 16.508 0.482 0.4891
anova(wm_all_AoLE, wm_all_AoLE_LoAE)
## Analysis of Variance Table
##
## Model 1: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Length.of.Auditory.Exposure
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 3544.2
## 2 103 3513.9 1 30.334 0.8892 0.3479
Checking assumptions for best-fit Working Memory model
ols_plot_resid_qq(wm_all_AoLE)
ols_test_normality(wm_all_AoLE)
## -----------------------------------------------
## Test Statistic pvalue
## -----------------------------------------------
## Shapiro-Wilk 0.9173 0.0000
## Kolmogorov-Smirnov 0.1184 0.0940
## Cramer-von Mises 9.025 0.0000
## Anderson-Darling 1.8261 1e-04
## -----------------------------------------------
ols_plot_resid_hist(wm_all_AoLE)
ols_test_correlation(wm_all_AoLE)
## [1] 0.9554127
ols_plot_resid_fit(wm_all_AoLE)
Step 1. Base Model:
plan_all_base <- lm(data=BRIEF_AgeOf, formula = Plan.Organize_RawScore~SES..3.66.+Sex+AgeMonths)
summary(plan_all_base)
##
## Call:
## lm(formula = Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.2655 -2.8630 -0.4688 2.5755 8.6681
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18.888939 2.203955 8.570 9.66e-14 ***
## SES..3.66. -0.076399 0.025890 -2.951 0.00391 **
## SexMale -0.224973 0.702342 -0.320 0.74936
## AgeMonths -0.003322 0.027010 -0.123 0.90234
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.602 on 105 degrees of freedom
## Multiple R-squared: 0.07663, Adjusted R-squared: 0.05024
## F-statistic: 2.904 on 3 and 105 DF, p-value: 0.03826
AIC(plan_all_base)
## [1] 594.6475
Step 2. Add Age of Language Exposure:
plan_all_AoLE <- lm(data=BRIEF_AgeOf, formula = Plan.Organize_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.)
summary(plan_all_AoLE)
##
## Call:
## lm(formula = Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.6566 -2.5795 -0.5434 2.0855 9.1255
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 19.36639 2.14243 9.039 9.37e-15 ***
## SES..3.66. -0.05741 0.02599 -2.209 0.02937 *
## SexMale -0.34626 0.68194 -0.508 0.61270
## AgeMonths -0.03909 0.02913 -1.342 0.18254
## Age.of.Language.Exposure..mo. 0.05740 0.02052 2.798 0.00613 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.491 on 104 degrees of freedom
## Multiple R-squared: 0.1413, Adjusted R-squared: 0.1082
## F-statistic: 4.277 on 4 and 104 DF, p-value: 0.003037
AIC(plan_all_AoLE)
## [1] 588.7381
Compare Model with Age of Language Exposure to base demographic model:
anova(plan_all_base, plan_all_AoLE)
## Analysis of Variance Table
##
## Model 1: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 1362.7
## 2 104 1267.3 1 95.378 7.8271 0.006134 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3a. Add Age of Auditory Experience:
plan_all_AoLE_AoAE <- lm(data=BRIEF_AgeOf, formula = Plan.Organize_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+AoAE)
summary(plan_all_AoLE_AoAE)
##
## Call:
## lm(formula = Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.3537 -2.3268 -0.6309 2.5441 8.6925
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 19.26342 2.13665 9.016 1.14e-14 ***
## SES..3.66. -0.06274 0.02622 -2.393 0.01853 *
## SexMale -0.31561 0.68005 -0.464 0.64355
## AgeMonths -0.02778 0.03029 -0.917 0.36136
## Age.of.Language.Exposure..mo. 0.06478 0.02122 3.054 0.00288 **
## AoAE -0.02023 0.01549 -1.306 0.19443
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.479 on 103 degrees of freedom
## Multiple R-squared: 0.1552, Adjusted R-squared: 0.1142
## F-statistic: 3.786 on 5 and 103 DF, p-value: 0.003455
AIC(plan_all_AoLE_AoAE)
## [1] 588.9476
3b. Add Length of Auditory Experience:
plan_all_AoLE_LoAE <- lm(data=BRIEF_AgeOf, formula = Plan.Organize_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Length.of.Auditory.Exposure)
summary(plan_all_AoLE_LoAE)
##
## Call:
## lm(formula = Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Length.of.Auditory.Exposure,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.3527 -2.3279 -0.6324 2.5437 8.6990
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 19.26082 2.13670 9.014 1.15e-14 ***
## SES..3.66. -0.06271 0.02622 -2.392 0.01858 *
## SexMale -0.31601 0.68003 -0.465 0.64313
## AgeMonths -0.04798 0.02982 -1.609 0.11063
## Age.of.Language.Exposure..mo. 0.06474 0.02120 3.053 0.00288 **
## Length.of.Auditory.Exposure 0.02024 0.01549 1.307 0.19420
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.479 on 103 degrees of freedom
## Multiple R-squared: 0.1553, Adjusted R-squared: 0.1143
## F-statistic: 3.786 on 5 and 103 DF, p-value: 0.003452
AIC(plan_all_AoLE_LoAE)
## [1] 588.9458
3c. Add Modality:
plan_all_AoLE_Modality <- lm(data=BRIEF_AgeOf, formula = Plan.Organize_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+Language_Modality)
summary(plan_all_AoLE_Modality)
##
## Call:
## lm(formula = Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Language_Modality, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.4928 -2.4645 -0.5671 2.2852 8.7057
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 19.65770 2.13218 9.220 4.02e-15 ***
## SES..3.66. -0.06371 0.02606 -2.445 0.01619 *
## SexMale -0.33942 0.67637 -0.502 0.61686
## AgeMonths -0.03294 0.02913 -1.131 0.26080
## Age.of.Language.Exposure..mo. 0.06720 0.02120 3.170 0.00201 **
## Language_ModalityASL -1.25648 0.76097 -1.651 0.10175
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.462 on 103 degrees of freedom
## Multiple R-squared: 0.1634, Adjusted R-squared: 0.1228
## F-statistic: 4.023 on 5 and 103 DF, p-value: 0.002244
AIC(plan_all_AoLE_Modality)
## [1] 587.8905
3d. Add Hearing Status (3-category):
plan_all_AoLE_HearStat <- lm(data=BRIEF_AgeOf, formula = Plan.Organize_RawScore~SES..3.66.+Sex+AgeMonths+Age.of.Language.Exposure..mo.+ Hearing_Level_3_Cat)
summary(plan_all_AoLE_HearStat)
##
## Call:
## lm(formula = Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths +
## Age.of.Language.Exposure..mo. + Hearing_Level_3_Cat, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.7011 -2.4923 -0.6156 2.1823 8.8189
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.02871 2.29794 8.716 8.5e-14 ***
## SES..3.66. -0.06211 0.02692 -2.307 0.02322 *
## SexMale -0.05949 0.70033 -0.085 0.93248
## AgeMonths -0.04095 0.03173 -1.290 0.20001
## Age.of.Language.Exposure..mo. 0.06796 0.02477 2.744 0.00725 **
## Hearing_Level_3_CatDeaf -1.14683 0.93669 -1.224 0.22382
## Hearing_Level_3_CatHard of Hearing -0.04387 1.19832 -0.037 0.97087
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.478 on 96 degrees of freedom
## (6 observations deleted due to missingness)
## Multiple R-squared: 0.1545, Adjusted R-squared: 0.1017
## F-statistic: 2.925 on 6 and 96 DF, p-value: 0.01156
AIC(plan_all_AoLE_HearStat)
## [1] 557.8476
Compare Models from Step 3a-c to model from Step 2:
anova(plan_all_AoLE, plan_all_AoLE_AoAE)
## Analysis of Variance Table
##
## Model 1: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1267.3
## 2 103 1246.7 1 20.647 1.7059 0.1944
anova(plan_all_AoLE, plan_all_AoLE_Modality)
## Analysis of Variance Table
##
## Model 1: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Language_Modality
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1267.3
## 2 103 1234.6 1 32.679 2.7263 0.1018
anova(plan_all_AoLE, plan_all_AoLE_LoAE)
## Analysis of Variance Table
##
## Model 1: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo.
## Model 2: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths + Age.of.Language.Exposure..mo. +
## Length.of.Auditory.Exposure
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1267.3
## 2 103 1246.6 1 20.668 1.7077 0.1942
Checking assumptions for best-fit Plan/Organize model
ols_plot_resid_qq(plan_all_AoLE)
ols_test_normality(plan_all_AoLE)
## -----------------------------------------------
## Test Statistic pvalue
## -----------------------------------------------
## Shapiro-Wilk 0.979 0.0831
## Kolmogorov-Smirnov 0.0865 0.3880
## Cramer-von Mises 9.0793 0.0000
## Anderson-Darling 0.7187 0.0591
## -----------------------------------------------
ols_plot_resid_hist(plan_all_AoLE)
ols_test_correlation(plan_all_AoLE)
## [1] 0.9908092
ols_plot_resid_fit(plan_all_AoLE)
stargazer(overall_all_AoLE_AoAE, inhibition_all_AoLE_AoAE, shift_all_AoLE_AoAE, emotctrl_all_AoLE_AoAE, wm_all_AoLE_AoAE, plan_all_AoLE_AoAE, type= "html", title = "Linear Regression Results for Raw Scores", align=TRUE, dep.var.labels=c("Global Executive Composite", "Inhibition", "Shift", "Emotional Control", "Working Memory", "Planning/Organization"), covariate.labels=c("Socioeconomic Status (SES)", "Sex (Male)", "Age (Months)", "Age of Exposure to Language (Months)", "Age of Auditory Exposure (Months)"), out="Table4.html")
This analysis looks at T-scores, and therefore only includes participants who were within the normed age range for the BRIEF-P
Create dataframe with participants who are age appropriate for the BRIEF-P (less than 5 years 11 Months old)
BRIEF_AgeApp <- subset(BRIEF, BRIEF$AgeMonths <= 71)
table1::label(BRIEF_AgeApp$AgeMonths) <- "Age (Months)"
table1::label(BRIEF_AgeApp$SES..3.66.) <- "SES"
table1::label(BRIEF_AgeApp$Sex) <- "Sex"
table1::label(BRIEF_AgeApp$Race_recoded) <- "Race"
table1::label(BRIEF_AgeApp$Ethnicity_recoded) <- "Ethnicity"
table1::label(BRIEF_AgeApp$Age.of.Language.Exposure..mo.) <- "Age of Language Exposure (Months)"
table1::label(BRIEF_AgeApp$AoAE) <- "Age of Auditory Exposure (Months)"
table1::table1(~AgeMonths + SES..3.66. + Sex + Race_recoded + Ethnicity_recoded + Age.of.Language.Exposure..mo. + AoAE, data = BRIEF_AgeApp)
| Overall (N=98) |
|
|---|---|
| Age (Months) | |
| Mean (SD) | 55.2 (9.87) |
| Median [Min, Max] | 56.0 [37.0, 71.0] |
| SES | |
| Mean (SD) | 51.3 (12.3) |
| Median [Min, Max] | 53.5 [9.00, 66.0] |
| Sex | |
| Female | 51 (52.0%) |
| Male | 47 (48.0%) |
| Race | |
| Asian | 1 (1.0%) |
| Black or African American | 1 (1.0%) |
| White | 87 (88.8%) |
| More than one | 7 (7.1%) |
| Other/Missing | 2 (2.0%) |
| Ethnicity | |
| Hispanic | 6 (6.1%) |
| Non-Hispanic | 78 (79.6%) |
| Prefer not to answer | 1 (1.0%) |
| Missing | 13 (13.3%) |
| Age of Language Exposure (Months) | |
| Mean (SD) | 9.17 (15.7) |
| Median [Min, Max] | 0 [0, 58.0] |
| Missing | 1 (1.0%) |
| Age of Auditory Exposure (Months) | |
| Mean (SD) | 15.1 (21.0) |
| Median [Min, Max] | 0.500 [0, 70.0] |
| Missing | 8 (8.2%) |
table1::table1(~AgeMonths + SES..3.66. + Sex + Race_recoded + Ethnicity_recoded + Age.of.Language.Exposure..mo. + AoAE | LanguageGroup, data = BRIEF_AgeApp, overall=F)
| Typically Hearing (N=44) |
Early ASL (N=20) |
Later English (N=20) |
Later ASL (N=14) |
|
|---|---|---|---|---|
| Age (Months) | ||||
| Mean (SD) | 53.7 (9.80) | 54.5 (8.98) | 58.3 (9.67) | 56.8 (11.3) |
| Median [Min, Max] | 54.0 [37.0, 71.0] | 54.5 [41.0, 71.0] | 60.0 [37.0, 71.0] | 60.0 [37.0, 71.0] |
| SES | ||||
| Mean (SD) | 55.5 (8.83) | 47.0 (16.0) | 49.6 (11.3) | 46.7 (14.2) |
| Median [Min, Max] | 56.0 [21.5, 66.0] | 53.8 [19.0, 66.0] | 50.5 [15.0, 66.0] | 50.8 [9.00, 62.0] |
| Sex | ||||
| Female | 23 (52.3%) | 11 (55.0%) | 10 (50.0%) | 7 (50.0%) |
| Male | 21 (47.7%) | 9 (45.0%) | 10 (50.0%) | 7 (50.0%) |
| Race | ||||
| Asian | 0 (0%) | 0 (0%) | 0 (0%) | 1 (7.1%) |
| Black or African American | 0 (0%) | 0 (0%) | 1 (5.0%) | 0 (0%) |
| White | 41 (93.2%) | 18 (90.0%) | 17 (85.0%) | 11 (78.6%) |
| More than one | 3 (6.8%) | 1 (5.0%) | 2 (10.0%) | 1 (7.1%) |
| Other/Missing | 0 (0%) | 1 (5.0%) | 0 (0%) | 1 (7.1%) |
| Ethnicity | ||||
| Hispanic | 3 (6.8%) | 0 (0%) | 1 (5.0%) | 2 (14.3%) |
| Non-Hispanic | 40 (90.9%) | 13 (65.0%) | 15 (75.0%) | 10 (71.4%) |
| Prefer not to answer | 0 (0%) | 1 (5.0%) | 0 (0%) | 0 (0%) |
| Missing | 1 (2.3%) | 6 (30.0%) | 4 (20.0%) | 2 (14.3%) |
| Age of Language Exposure (Months) | ||||
| Mean (SD) | 0 (0) | 0 (0) | 22.0 (17.2) | 34.6 (8.23) |
| Median [Min, Max] | 0 [0, 0] | 0 [0, 0] | 20.0 [0.500, 58.0] | 36.0 [18.0, 46.0] |
| Missing | 0 (0%) | 0 (0%) | 0 (0%) | 1 (7.1%) |
| Age of Auditory Exposure (Months) | ||||
| Mean (SD) | 0 (0) | 44.4 (20.9) | 22.0 (17.2) | 25.2 (18.3) |
| Median [Min, Max] | 0 [0, 0] | 46.5 [1.00, 70.0] | 20.0 [0, 58.0] | 27.0 [3.00, 60.0] |
| Missing | 0 (0%) | 6 (30.0%) | 0 (0%) | 2 (14.3%) |
G <- ggplot(data=BRIEF_AgeApp, mapping = aes(x= Language_Timing, y=GEC_Tscore)) + geom_boxplot() + geom_dotplot(mapping=aes(fill=Language_Modality), dotsize = 1.1, method = "dotdensity", binaxis = "y", stackdir = "center", alpha=0.8) + theme(text = element_text(size=20), axis.title=element_text(size=20,face="bold"), legend.title=element_text(size=24), legend.text=element_text(size=24), axis.title.x=element_blank(), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank(), axis.line = element_line(colour = "black")) + ylab("Global Executive Composite (T-scores)") + scale_x_discrete(labels=c("Early" = "Early Language", "Later" = "Later Language")) + scale_y_continuous(limits= c(30, 100), breaks= c(30, 40, 50, 60, 70, 80, 90, 100)) + scale_fill_grey(name = "Language Modality", start=0.1, end=0.9, ) + geom_hline(aes(yintercept= 50, linetype = "Mean"), colour= 'black') + geom_hline(aes(yintercept= 65, linetype = "Elevated"), colour= 'grey17') + scale_linetype_manual(name = "Standard Score", values=c("dotted", "longdash"), guide = guide_legend(override.aes = list(color = c("grey17", "black"))))
I <- ggplot(data=BRIEF_AgeApp, mapping = aes(x= Language_Timing, y=Inhibit_Tscore)) + geom_boxplot() + geom_dotplot(mapping=aes(fill=Language_Modality), dotsize = 1.1, method = "dotdensity", binaxis = "y", stackdir = "center", alpha=0.8) + theme(text = element_text(size=20), axis.title=element_text(size=20,face="bold"), legend.title=element_text(size=24), legend.text=element_text(size=24), axis.title.x=element_blank(), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank(), axis.line = element_line(colour = "black")) + ylab("Inhibition (T-scores)") + scale_x_discrete(labels=c("Early" = "Early Language", "Later" = "Later Language")) + scale_y_continuous(limits= c(30, 100), breaks= c(30, 40, 50, 60, 70, 80, 90, 100)) + scale_fill_grey(name = "Language Modality", start=0.1, end=0.9) + geom_hline(aes(yintercept= 50, linetype = "Mean"), colour= 'black') + geom_hline(aes(yintercept= 65, linetype = "Elevated"), colour= 'grey17') + scale_linetype_manual(name = "Standard Score", values=c("dotted", "longdash"), guide = guide_legend(override.aes = list(color = c("grey17", "black"))))
S <- ggplot(data=BRIEF_AgeApp, mapping = aes(x= Language_Timing, y=Shift_Tscore)) + geom_boxplot() + geom_dotplot(mapping=aes(fill=Language_Modality), dotsize = 1.1, method = "dotdensity", binaxis = "y", stackdir = "center", alpha=0.8) + theme(text = element_text(size=20), axis.title=element_text(size=20,face="bold"), legend.title=element_text(size=24), legend.text=element_text(size=24), axis.title.x=element_blank(), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank(), axis.line = element_line(colour = "black")) + ylab("Shift (T-scores)") + scale_x_discrete(labels=c("Early" = "Early Language", "Later" = "Later Language")) + scale_y_continuous(limits= c(30, 100), breaks= c(30, 40, 50, 60, 70, 80, 90, 100)) + scale_fill_grey(name = "Language Modality", start=0.1, end=0.9) + geom_hline(aes(yintercept= 50, linetype = "Mean"), colour= 'black') + geom_hline(aes(yintercept= 65, linetype = "Elevated"), colour= 'grey17') + scale_linetype_manual(name = "Standard Score", values=c("dotted", "longdash"), guide = guide_legend(override.aes = list(color = c("grey17", "black"))))
E <- ggplot(data=BRIEF_AgeApp, mapping = aes(x= Language_Timing, y= Emotional.Control_Tscore)) + geom_boxplot() + geom_dotplot(mapping=aes(fill=Language_Modality), dotsize = 1.1, method = "dotdensity", binaxis = "y", stackdir = "center", alpha=0.8) + theme(text = element_text(size=20), axis.title=element_text(size=20,face="bold"), legend.title=element_text(size=24), legend.text=element_text(size=24), axis.title.x=element_blank(), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank(), axis.line = element_line(colour = "black")) + ylab("Emotional Control (T-scores)") + scale_x_discrete(labels=c("Early" = "Early Language", "Later" = "Later Language")) + scale_y_continuous(limits= c(30, 100), breaks= c(30, 40, 50, 60, 70, 80, 90, 100)) + scale_fill_grey(name = "Language Modality", start=0.1, end=0.9) + geom_hline(aes(yintercept= 50, linetype = "Mean"), colour= 'black') + geom_hline(aes(yintercept= 65, linetype = "Elevated"), colour= 'grey17') + scale_linetype_manual(name = "Standard Score", values=c("dotted", "longdash"), guide = guide_legend(override.aes = list(color = c("grey17", "black"))))
W <- ggplot(data=BRIEF_AgeApp, mapping = aes(x= Language_Timing, y= Working.Memory_Tscore)) + geom_boxplot() + geom_dotplot(mapping=aes(fill=Language_Modality), dotsize = 1.1, method = "dotdensity", binaxis = "y", stackdir = "center", alpha=0.8) + theme(text = element_text(size=20), axis.title=element_text(size=20,face="bold"), legend.title=element_text(size=24), legend.text=element_text(size=24), axis.title.x=element_blank(), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank(), axis.line = element_line(colour = "black")) + ylab("Working Memory (T-scores)") + scale_x_discrete(labels=c("Early" = "Early Language", "Later" = "Later Language")) + scale_y_continuous(limits= c(30, 100), breaks= c(30, 40, 50, 60, 70, 80, 90, 100)) + scale_fill_grey(name = "Language Modality", start=0.1, end=0.9) + geom_hline(aes(yintercept= 50, linetype = "Mean"), colour= 'black') + geom_hline(aes(yintercept= 65, linetype = "Elevated"), colour= 'grey17') + scale_linetype_manual(name = "Standard Score", values=c("dotted", "longdash"), guide = guide_legend(override.aes = list(color = c("grey17", "black"))))
P <- ggplot(data=BRIEF_AgeApp, mapping = aes(x= Language_Timing, y= Plan.Organize_Tscore)) + geom_boxplot() + geom_dotplot(mapping=aes(fill=Language_Modality), dotsize = 1.1, method = "dotdensity", binaxis = "y", stackdir = "center", alpha=0.8) + theme(text = element_text(size=20), axis.title=element_text(size=20,face="bold"), legend.title=element_text(size=24), legend.text=element_text(size=24), axis.title.x=element_blank(), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank(), axis.line = element_line(colour = "black")) + ylab("Plan/Organize (T-scores)") + scale_x_discrete(labels=c("Early" = "Early Language", "Later" = "Later Language")) + scale_y_continuous(limits= c(30, 100), breaks= c(30, 40, 50, 60, 70, 80, 90, 100)) + scale_fill_grey(name = "Language Modality", start=0.1, end=0.9) + geom_hline(aes(yintercept= 50, linetype = "Mean"), colour= 'black') + geom_hline(aes(yintercept= 65, linetype = "Elevated"), colour= 'grey17') + scale_linetype_manual(name = "Standard Score", values=c("dotted", "longdash"), guide = guide_legend(override.aes = list(color = c("grey17", "black"))))
ggarrange(G, I, S, E, W, P, ncol=3, nrow=2, common.legend = TRUE, legend="bottom")
BRIEF_AgeApp_GEC_over65 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Global Executive Composite T-score < 65 (n)" = sum(GEC_Tscore < 65), "Global Executive Composite T-score > 65 (n)" = sum(GEC_Tscore >= 65))
BRIEF_AgeApp_GEC_over65 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Global Executive Composite T-score < 65 (n) | Global Executive Composite T-score > 65 (n) |
|---|---|---|
| Typically Hearing | 41 | 3 |
| Early ASL | 19 | 1 |
| Later English | 17 | 3 |
| Later ASL | 12 | 2 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
GEC_rr <- matrix(c(41, 3, 19, 1, 17, 3, 12, 2), 4, 2, byrow=TRUE)
dimnames(GEC_rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "GEC T-score" = c("Not Clinically Significant", "Clinically Significant"))
GEC_rr
## GEC T-score
## Group Not Clinically Significant Clinically Significant
## Typically Hearing 41 3
## Early ASL 19 1
## Later English 17 3
## Later ASL 12 2
riskratio.small(GEC_rr, verbose=TRUE)
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## $x
## GEC T-score
## Group Not Clinically Significant Clinically Significant
## Typically Hearing 41 3
## Early ASL 19 1
## Later English 17 3
## Later ASL 12 2
##
## $data
## GEC T-score
## Group Not Clinically Significant Clinically Significant Total
## Typically Hearing 41 3 44
## Early ASL 19 1 20
## Later English 17 3 20
## Later ASL 12 2 14
## Total 89 9 98
##
## $p.exposed
## GEC T-score
## Group Not Clinically Significant Clinically Significant Total
## Typically Hearing 0.4606742 0.3333333 0.4489796
## Early ASL 0.2134831 0.1111111 0.2040816
## Later English 0.1910112 0.3333333 0.2040816
## Later ASL 0.1348315 0.2222222 0.1428571
## Total 1.0000000 1.0000000 1.0000000
##
## $p.outcome
## GEC T-score
## Group Not Clinically Significant Clinically Significant Total
## Typically Hearing 0.9318182 0.06818182 1
## Early ASL 0.9500000 0.05000000 1
## Later English 0.8500000 0.15000000 1
## Later ASL 0.8571429 0.14285714 1
## Total 0.9081633 0.09183673 1
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.000000 NA NA
## Early ASL 0.562500 0.06229044 5.079531
## Later English 1.687500 0.37260128 7.642637
## Later ASL 1.607143 0.29800962 8.667197
##
## $conf.level
## [1] 0.95
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.8441962 1.0000000 0.7806100
## Later English 0.3428741 0.3662788 0.2979414
## Later ASL 0.4334207 0.5852322 0.3858989
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
BRIEF_AgeApp_Inhibit_over65 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Inhibit T-score < 65 (n)" = sum(Inhibit_Tscore < 65), "Inhibit T-score > 65 (n)" = sum(Inhibit_Tscore >= 65))
BRIEF_AgeApp_Inhibit_over65 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Inhibit T-score < 65 (n) | Inhibit T-score > 65 (n) |
|---|---|---|
| Typically Hearing | 42 | 2 |
| Early ASL | 18 | 2 |
| Later English | 19 | 1 |
| Later ASL | 14 | 0 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
Inhibit_rr <- matrix(c(42,2, 18, 2, 19, 1, 14, 0), 4, 2, byrow=TRUE)
dimnames(Inhibit_rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "GEC T-score" = c("Not Clinically Significant", "Clinically Significant"))
Inhibit_rr
## GEC T-score
## Group Not Clinically Significant Clinically Significant
## Typically Hearing 42 2
## Early ASL 18 2
## Later English 19 1
## Later ASL 14 0
riskratio.small(Inhibit_rr)
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## $data
## GEC T-score
## Group Not Clinically Significant Clinically Significant Total
## Typically Hearing 42 2 44
## Early ASL 18 2 20
## Later English 19 1 20
## Later ASL 14 0 14
## Total 93 5 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.00 NA NA
## Early ASL 1.50 0.22720962 9.902750
## Later English 0.75 0.07213456 7.797927
## Later ASL 0.00 0.00000000 NaN
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.4560292 0.583113 0.4033953
## Later English 0.9101382 1.000000 0.9364430
## Later ASL 0.5722928 1.000000 0.4168811
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
BRIEF_AgeApp_Shift_over65 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Shift T-score < 65 (n)" = sum(Shift_Tscore < 65), "Shift T-score > 65 (n)" = sum(Shift_Tscore >= 65))
BRIEF_AgeApp_Shift_over65 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Shift T-score < 65 (n) | Shift T-score > 65 (n) |
|---|---|---|
| Typically Hearing | 42 | 2 |
| Early ASL | 19 | 1 |
| Later English | 18 | 2 |
| Later ASL | 13 | 1 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
Shift_rr <- matrix(c(42,2,19,1, 18, 2, 13, 1), 4, 2, byrow=TRUE)
dimnames(Shift_rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "Shift T-score" = c("Not Clinically Significant", "Clinically Significant"))
Shift_rr
## Shift T-score
## Group Not Clinically Significant Clinically Significant
## Typically Hearing 42 2
## Early ASL 19 1
## Later English 18 2
## Later ASL 13 1
riskratio.small(Shift_rr)
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## $data
## Shift T-score
## Group Not Clinically Significant Clinically Significant Total
## Typically Hearing 42 2 44
## Early ASL 19 1 20
## Later English 18 2 20
## Later ASL 13 1 14
## Total 92 6 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.000000 NA NA
## Early ASL 0.750000 0.07213456 7.797927
## Later English 1.500000 0.22720962 9.902750
## Later ASL 1.071429 0.10488371 10.945067
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.9101382 1.000000 0.9364430
## Later English 0.4560292 0.583113 0.4033953
## Later ASL 0.7123412 1.000000 0.7023059
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
BRIEF_AgeApp_EmoCtrl_over65 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Emotional Control T-score < 65 (n)" = sum(Emotional.Control_Tscore < 65), "Emotional Control T-score > 65 (n)" = sum(Emotional.Control_Tscore >= 65))
BRIEF_AgeApp_EmoCtrl_over65 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Emotional Control T-score < 65 (n) | Emotional Control T-score > 65 (n) |
|---|---|---|
| Typically Hearing | 41 | 3 |
| Early ASL | 20 | 0 |
| Later English | 18 | 2 |
| Later ASL | 12 | 2 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
EmoCtrl_rr <- matrix(c(41,3,20,0, 18, 2, 12, 2), 4, 2, byrow=TRUE)
dimnames(EmoCtrl_rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "Emotional Control T-score" = c("Not Clinically Significant", "Clinically Significant"))
EmoCtrl_rr
## Emotional Control T-score
## Group Not Clinically Significant Clinically Significant
## Typically Hearing 41 3
## Early ASL 20 0
## Later English 18 2
## Later ASL 12 2
riskratio.small(EmoCtrl_rr)
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## $data
## Emotional Control T-score
## Group Not Clinically Significant Clinically Significant Total
## Typically Hearing 41 3 44
## Early ASL 20 0 20
## Later English 18 2 20
## Later ASL 12 2 14
## Total 91 7 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.000000 NA NA
## Early ASL 0.000000 0.0000000 NaN
## Later English 1.125000 0.2036087 6.215966
## Later ASL 1.607143 0.2980096 8.667197
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.3178763 0.5458909 0.2316502
## Later English 0.6729097 0.6439090 0.6601990
## Later ASL 0.4334207 0.5852322 0.3858989
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
BRIEF_AgeApp_WorkMem_over65 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Working Memory T-score < 65 (n)" = sum(Working.Memory_Tscore < 65), "Working Memory T-score > 65 (n)" = sum(Working.Memory_Tscore >= 65))
BRIEF_AgeApp_WorkMem_over65 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Working Memory T-score < 65 (n) | Working Memory T-score > 65 (n) |
|---|---|---|
| Typically Hearing | 39 | 5 |
| Early ASL | 17 | 3 |
| Later English | 18 | 2 |
| Later ASL | 13 | 1 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
WorkMem_rr <- matrix(c(39,5,17,3, 18, 2, 13, 1), 4, 2, byrow=TRUE)
dimnames(WorkMem_rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "Working Memory T-score" = c("Not Clinically Significant", "Clinically Significant"))
WorkMem_rr
## Working Memory T-score
## Group Not Clinically Significant Clinically Significant
## Typically Hearing 39 5
## Early ASL 17 3
## Later English 18 2
## Later ASL 13 1
riskratio.small(WorkMem_rr)
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## $data
## Working Memory T-score
## Group Not Clinically Significant Clinically Significant Total
## Typically Hearing 39 5 44
## Early ASL 17 3 20
## Later English 18 2 20
## Later ASL 13 1 14
## Total 87 11 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.0000000 NA NA
## Early ASL 1.1250000 0.29748109 4.254472
## Later English 0.7500000 0.15881991 3.541747
## Later ASL 0.5357143 0.06820514 4.207744
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.6878536 0.6969793 0.6834809
## Later English 0.9100612 1.0000000 0.8712975
## Later ASL 0.7244461 1.0000000 0.6515079
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
BRIEF_AgeApp_PlanOrg_over65 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Plan/Organize T-score < 65 (n)" = sum(Plan.Organize_Tscore < 65), "Plan/Organize T-score > 65 (n)" = sum(Plan.Organize_Tscore >= 65))
BRIEF_AgeApp_PlanOrg_over65 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Plan/Organize T-score < 65 (n) | Plan/Organize T-score > 65 (n) |
|---|---|---|
| Typically Hearing | 41 | 3 |
| Early ASL | 19 | 1 |
| Later English | 16 | 4 |
| Later ASL | 14 | 0 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
PlanOrg_rr <- matrix(c(41,3,19,1, 16, 4, 14, 0), 4, 2, byrow=TRUE)
dimnames(PlanOrg_rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "Plan/Organize T-score" = c("Not Clinically Significant", "Clinically Significant"))
PlanOrg_rr
## Plan/Organize T-score
## Group Not Clinically Significant Clinically Significant
## Typically Hearing 41 3
## Early ASL 19 1
## Later English 16 4
## Later ASL 14 0
riskratio.small(PlanOrg_rr)
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## Warning in chisq.test(xx, correct = correction): Chi-squared approximation may
## be incorrect
## $data
## Plan/Organize T-score
## Group Not Clinically Significant Clinically Significant Total
## Typically Hearing 41 3 44
## Early ASL 19 1 20
## Later English 16 4 20
## Later ASL 14 0 14
## Total 90 8 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.0000 NA NA
## Early ASL 0.5625 0.06229044 5.079531
## Later English 2.2500 0.55455158 9.128998
## Later ASL 0.0000 0.00000000 NaN
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.8441962 1.0000000 0.7806100
## Later English 0.1562526 0.1914591 0.1173246
## Later ASL 0.4292196 1.0000000 0.3157160
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
Possible Issue: Because Age of Language Exposure and Age of Auditory Exposure were significantly correlated, we need to determine whether they might actually be measuring same underlying construct
Approach: See whether findings above (where Language but not Auditory exposure significantly predicts BRIEF scores) hold when predictors are added to the model in a different order
overall_all_base <- lm(data=BRIEF_AgeOf, formula = GEC_RawScore~SES..3.66.+Sex+AgeMonths)
summary(overall_all_base)
##
## Call:
## lm(formula = GEC_RawScore ~ SES..3.66. + Sex + AgeMonths, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -32.330 -14.855 -0.360 8.628 60.531
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 107.1580 11.5903 9.245 3.02e-15 ***
## SES..3.66. -0.4345 0.1362 -3.191 0.00187 **
## SexMale -1.1793 3.6935 -0.319 0.75014
## AgeMonths 0.1167 0.1420 0.822 0.41320
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 18.94 on 105 degrees of freedom
## Multiple R-squared: 0.09454, Adjusted R-squared: 0.06867
## F-statistic: 3.654 on 3 and 105 DF, p-value: 0.01492
overall_all_AoAE <- lm(data=BRIEF_AgeOf, formula = GEC_RawScore~SES..3.66.+Sex+AgeMonths+AoAE)
summary(overall_all_AoAE)
##
## Call:
## lm(formula = GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -32.739 -15.336 -0.297 8.446 60.008
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 107.00901 11.66195 9.176 4.66e-15 ***
## SES..3.66. -0.44157 0.14033 -3.147 0.00215 **
## SexMale -1.13741 3.71503 -0.306 0.76009
## AgeMonths 0.13112 0.15648 0.838 0.40400
## AoAE -0.01834 0.08166 -0.225 0.82276
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 19.03 on 104 degrees of freedom
## Multiple R-squared: 0.09498, Adjusted R-squared: 0.06017
## F-statistic: 2.729 on 4 and 104 DF, p-value: 0.03311
anova(overall_all_base, overall_all_AoAE)
## Analysis of Variance Table
##
## Model 1: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 37686
## 2 104 37668 1 18.264 0.0504 0.8228
overall_all_AoAE_AoLE <- lm(data=BRIEF_AgeOf, formula = GEC_RawScore~SES..3.66.+Sex+AgeMonths+AoAE+Age.of.Language.Exposure..mo.)
summary(overall_all_AoAE_AoLE)
##
## Call:
## lm(formula = GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE +
## Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -35.488 -13.012 -0.955 9.269 64.745
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 109.30858 11.24579 9.720 3.11e-16 ***
## SES..3.66. -0.35417 0.13801 -2.566 0.01172 *
## SexMale -1.70684 3.57928 -0.477 0.63447
## AgeMonths -0.02933 0.15945 -0.184 0.85440
## AoAE -0.08470 0.08152 -1.039 0.30121
## Age.of.Language.Exposure..mo. 0.34130 0.11166 3.057 0.00285 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 18.31 on 103 degrees of freedom
## Multiple R-squared: 0.1702, Adjusted R-squared: 0.13
## F-statistic: 4.226 on 5 and 103 DF, p-value: 0.001553
anova(overall_all_AoAE, overall_all_AoAE_AoLE)
## Analysis of Variance Table
##
## Model 1: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Model 2: GEC_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE + Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 37668
## 2 103 34535 1 3132.4 9.3425 0.002852 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#~~~~~#
inhibition_all_base <- lm(data=BRIEF_AgeOf, formula = Inhibit_RawScore~SES..3.66.+Sex+AgeMonths)
summary(inhibition_all_base)
##
## Call:
## lm(formula = Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.4961 -4.5032 0.0108 4.0216 21.5107
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.07731 3.65405 6.863 4.86e-10 ***
## SES..3.66. -0.10990 0.04292 -2.560 0.0119 *
## SexMale 0.97828 1.16445 0.840 0.4028
## AgeMonths 0.06696 0.04478 1.495 0.1378
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.973 on 105 degrees of freedom
## Multiple R-squared: 0.08539, Adjusted R-squared: 0.05926
## F-statistic: 3.268 on 3 and 105 DF, p-value: 0.02424
inhibition_all_AoAE <- lm(data=BRIEF_AgeOf, formula = Inhibit_RawScore~SES..3.66.+Sex+AgeMonths+AoAE)
summary(inhibition_all_AoAE)
##
## Call:
## lm(formula = Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths +
## AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.2413 -4.6083 0.1954 4.1177 21.2498
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.002918 3.675294 6.803 6.7e-10 ***
## SES..3.66. -0.113419 0.044224 -2.565 0.0118 *
## SexMale 0.999205 1.170801 0.853 0.3954
## AgeMonths 0.074163 0.049315 1.504 0.1357
## AoAE -0.009156 0.025737 -0.356 0.7227
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.998 on 104 degrees of freedom
## Multiple R-squared: 0.0865, Adjusted R-squared: 0.05137
## F-statistic: 2.462 on 4 and 104 DF, p-value: 0.04977
anova(inhibition_all_base, inhibition_all_AoAE)
## Analysis of Variance Table
##
## Model 1: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 3745.7
## 2 104 3741.2 1 4.5532 0.1266 0.7227
inhibition_all_AoAE_AoLE <- lm(data=BRIEF_AgeOf, formula = Inhibit_RawScore~SES..3.66.+Sex+AgeMonths+AoAE+Age.of.Language.Exposure..mo.)
summary(inhibition_all_AoAE_AoLE)
##
## Call:
## lm(formula = Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths +
## AoAE + Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.3417 -4.3176 -0.1074 4.1316 22.2424
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.48478 3.63271 7.015 2.49e-10 ***
## SES..3.66. -0.09511 0.04458 -2.133 0.0353 *
## SexMale 0.87988 1.15621 0.761 0.4484
## AgeMonths 0.04054 0.05151 0.787 0.4330
## AoAE -0.02306 0.02633 -0.876 0.3832
## Age.of.Language.Exposure..mo. 0.07152 0.03607 1.983 0.0501 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.915 on 103 degrees of freedom
## Multiple R-squared: 0.1201, Adjusted R-squared: 0.07737
## F-statistic: 2.811 on 5 and 103 DF, p-value: 0.02018
anova(inhibition_all_AoAE, inhibition_all_AoAE_AoLE)
## Analysis of Variance Table
##
## Model 1: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Model 2: Inhibit_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE + Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 3741.2
## 2 103 3603.6 1 137.54 3.9313 0.05006 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#~~~~~#
shift_all_base <- lm(data=BRIEF_AgeOf, formula = Shift_RawScore~SES..3.66.+Sex+AgeMonths)
summary(shift_all_base)
##
## Call:
## lm(formula = Shift_RawScore ~ SES..3.66. + Sex + AgeMonths, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.936 -2.682 -0.776 1.774 14.985
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 17.93727 2.20373 8.140 8.65e-13 ***
## SES..3.66. -0.08380 0.02589 -3.237 0.00162 **
## SexMale -0.79353 0.70227 -1.130 0.26107
## AgeMonths 0.01077 0.02701 0.399 0.69088
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.602 on 105 degrees of freedom
## Multiple R-squared: 0.0979, Adjusted R-squared: 0.07213
## F-statistic: 3.799 on 3 and 105 DF, p-value: 0.01245
shift_all_AoAE <- lm(data=BRIEF_AgeOf, formula = Shift_RawScore~SES..3.66.+Sex+AgeMonths+AoAE)
summary(shift_all_AoAE)
##
## Call:
## lm(formula = Shift_RawScore ~ SES..3.66. + Sex + AgeMonths +
## AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.7838 -2.6726 -0.7207 1.9186 14.8359
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18.012293 2.214116 8.135 9.32e-13 ***
## SES..3.66. -0.080255 0.026642 -3.012 0.00326 **
## SexMale -0.814632 0.705328 -1.155 0.25075
## AgeMonths 0.003507 0.029709 0.118 0.90625
## AoAE 0.009233 0.015505 0.596 0.55279
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.613 on 104 degrees of freedom
## Multiple R-squared: 0.101, Adjusted R-squared: 0.06639
## F-statistic: 2.92 on 4 and 104 DF, p-value: 0.02467
anova(shift_all_base, shift_all_AoAE)
## Analysis of Variance Table
##
## Model 1: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 1362.4
## 2 104 1357.8 1 4.63 0.3546 0.5528
shift_all_AoAE_AoLE <- lm(data=BRIEF_AgeOf, formula = Shift_RawScore~SES..3.66.+Sex+AgeMonths+AoAE+Age.of.Language.Exposure..mo.)
summary(shift_all_AoAE_AoLE)
##
## Call:
## lm(formula = Shift_RawScore ~ SES..3.66. + Sex + AgeMonths +
## AoAE + Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.4369 -2.2772 -0.5143 1.4576 14.0693
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18.425943 2.144990 8.590 9.89e-14 ***
## SES..3.66. -0.064533 0.026323 -2.452 0.0159 *
## SexMale -0.917061 0.682702 -1.343 0.1821
## AgeMonths -0.025355 0.030413 -0.834 0.4064
## AoAE -0.002704 0.015549 -0.174 0.8623
## Age.of.Language.Exposure..mo. 0.061394 0.021298 2.883 0.0048 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.493 on 103 degrees of freedom
## Multiple R-squared: 0.1681, Adjusted R-squared: 0.1277
## F-statistic: 4.162 on 5 and 103 DF, p-value: 0.001745
anova(shift_all_AoAE, shift_all_AoAE_AoLE)
## Analysis of Variance Table
##
## Model 1: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Model 2: Shift_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE + Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1357.8
## 2 103 1256.4 1 101.36 8.3093 0.004802 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#~~~~~#
emotctrl_all_base <- lm(data=BRIEF_AgeOf, formula = Emotional.Control_RawScore~SES..3.66.+Sex+AgeMonths)
emotctrl_all_AoAE <- lm(data=BRIEF_AgeOf, formula = Emotional.Control_RawScore~SES..3.66.+Sex+AgeMonths+AoAE)
summary(emotctrl_all_AoAE)
##
## Call:
## lm(formula = Emotional.Control_RawScore ~ SES..3.66. + Sex +
## AgeMonths + AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.6343 -3.0275 -0.6445 2.5623 13.9640
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 19.986525 2.512209 7.956 2.3e-12 ***
## SES..3.66. -0.094341 0.030229 -3.121 0.00233 **
## SexMale -0.688072 0.800289 -0.860 0.39189
## AgeMonths 0.007952 0.033709 0.236 0.81396
## AoAE -0.007747 0.017592 -0.440 0.66059
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.1 on 104 degrees of freedom
## Multiple R-squared: 0.08881, Adjusted R-squared: 0.05376
## F-statistic: 2.534 on 4 and 104 DF, p-value: 0.04459
anova(emotctrl_all_base, emotctrl_all_AoAE)
## Analysis of Variance Table
##
## Model 1: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 1751.2
## 2 104 1748.0 1 3.2593 0.1939 0.6606
emotctrl_all_AoAE_AoLE <- lm(data=BRIEF_AgeOf, formula = Emotional.Control_RawScore~SES..3.66.+Sex+AgeMonths+AoAE+Age.of.Language.Exposure..mo.)
summary(emotctrl_all_AoAE_AoLE)
##
## Call:
## lm(formula = Emotional.Control_RawScore ~ SES..3.66. + Sex +
## AgeMonths + AoAE + Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.418 -3.282 -0.533 2.551 13.559
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.204803 2.509534 8.051 1.5e-12 ***
## SES..3.66. -0.086045 0.030797 -2.794 0.00621 **
## SexMale -0.742122 0.798728 -0.929 0.35499
## AgeMonths -0.007278 0.035581 -0.205 0.83833
## AoAE -0.014046 0.018191 -0.772 0.44179
## Age.of.Language.Exposure..mo. 0.032397 0.024918 1.300 0.19645
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.086 on 103 degrees of freedom
## Multiple R-squared: 0.1035, Adjusted R-squared: 0.06
## F-statistic: 2.379 on 5 and 103 DF, p-value: 0.0437
anova(emotctrl_all_AoAE, emotctrl_all_AoAE_AoLE)
## Analysis of Variance Table
##
## Model 1: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Model 2: Emotional.Control_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE +
## Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1748.0
## 2 103 1719.8 1 28.224 1.6904 0.1965
#~~~~~#
wm_all_base <- lm(data=BRIEF_AgeOf, formula = Working.Memory_RawScore~SES..3.66.+Sex+AgeMonths)
summary(wm_all_base)
##
## Call:
## lm(formula = Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.748 -4.607 -1.414 3.432 23.472
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.20501 3.70291 6.807 6.38e-10 ***
## SES..3.66. -0.07306 0.04350 -1.680 0.096 .
## SexMale -0.43333 1.18002 -0.367 0.714
## AgeMonths 0.04043 0.04538 0.891 0.375
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 6.053 on 105 degrees of freedom
## Multiple R-squared: 0.0346, Adjusted R-squared: 0.007015
## F-statistic: 1.254 on 3 and 105 DF, p-value: 0.294
wm_all_AoAE <- lm(data=BRIEF_AgeOf, formula = Working.Memory_RawScore~SES..3.66.+Sex+AgeMonths+AoAE)
summary(wm_all_AoAE)
##
## Call:
## lm(formula = Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths +
## AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.664 -4.636 -1.421 3.421 23.385
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.180342 3.726467 6.757 8.35e-10 ***
## SES..3.66. -0.074223 0.044840 -1.655 0.101
## SexMale -0.426386 1.187103 -0.359 0.720
## AgeMonths 0.042816 0.050002 0.856 0.394
## AoAE -0.003036 0.026095 -0.116 0.908
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 6.081 on 104 degrees of freedom
## Multiple R-squared: 0.03472, Adjusted R-squared: -0.002403
## F-statistic: 0.9353 on 4 and 104 DF, p-value: 0.4466
anova(wm_all_base, wm_all_AoAE)
## Analysis of Variance Table
##
## Model 1: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 3846.6
## 2 104 3846.1 1 0.50068 0.0135 0.9076
wm_all_AoAE_AoLE <- lm(data=BRIEF_AgeOf, formula = Working.Memory_RawScore~SES..3.66.+Sex+AgeMonths+AoAE+Age.of.Language.Exposure..mo.)
summary(wm_all_AoAE_AoLE)
##
## Call:
## lm(formula = Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths +
## AoAE + Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -9.313 -4.115 -1.080 3.208 24.929
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.929624 3.586996 7.229 8.82e-11 ***
## SES..3.66. -0.045745 0.044020 -1.039 0.30115
## SexMale -0.611926 1.141660 -0.536 0.59312
## AgeMonths -0.009465 0.050858 -0.186 0.85272
## AoAE -0.024660 0.026002 -0.948 0.34514
## Age.of.Language.Exposure..mo. 0.111208 0.035616 3.122 0.00233 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.841 on 103 degrees of freedom
## Multiple R-squared: 0.1182, Adjusted R-squared: 0.07538
## F-statistic: 2.761 on 5 and 103 DF, p-value: 0.02209
anova(wm_all_AoAE, wm_all_AoAE_AoLE)
## Analysis of Variance Table
##
## Model 1: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Model 2: Working.Memory_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE +
## Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 3846.1
## 2 103 3513.5 1 332.57 9.7494 0.002329 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#~~~~~#
plan_all_base <- lm(data=BRIEF_AgeOf, formula = Plan.Organize_RawScore~SES..3.66.+Sex+AgeMonths)
summary(plan_all_base)
##
## Call:
## lm(formula = Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths,
## data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.2655 -2.8630 -0.4688 2.5755 8.6681
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18.888939 2.203955 8.570 9.66e-14 ***
## SES..3.66. -0.076399 0.025890 -2.951 0.00391 **
## SexMale -0.224973 0.702342 -0.320 0.74936
## AgeMonths -0.003322 0.027010 -0.123 0.90234
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.602 on 105 degrees of freedom
## Multiple R-squared: 0.07663, Adjusted R-squared: 0.05024
## F-statistic: 2.904 on 3 and 105 DF, p-value: 0.03826
plan_all_AoAE <- lm(data=BRIEF_AgeOf, formula = Plan.Organize_RawScore~SES..3.66.+Sex+AgeMonths+AoAE)
summary(plan_all_AoAE)
##
## Call:
## lm(formula = Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths +
## AoAE, data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.2075 -2.9021 -0.5068 2.7758 8.5542
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18.826930 2.215542 8.498 1.49e-13 ***
## SES..3.66. -0.079332 0.026659 -2.976 0.00363 **
## SexMale -0.207528 0.705783 -0.294 0.76931
## AgeMonths 0.002680 0.029728 0.090 0.92833
## AoAE -0.007632 0.015515 -0.492 0.62380
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.616 on 104 degrees of freedom
## Multiple R-squared: 0.07877, Adjusted R-squared: 0.04334
## F-statistic: 2.223 on 4 and 104 DF, p-value: 0.07151
anova(plan_all_base, plan_all_AoAE)
## Analysis of Variance Table
##
## Model 1: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths
## Model 2: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 105 1362.7
## 2 104 1359.5 1 3.1635 0.242 0.6238
plan_all_AoAE_AoLE <- lm(data=BRIEF_AgeOf, formula = Plan.Organize_RawScore~SES..3.66.+Sex+AgeMonths+AoAE+Age.of.Language.Exposure..mo.)
summary(plan_all_AoAE_AoLE)
##
## Call:
## lm(formula = Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths +
## AoAE + Age.of.Language.Exposure..mo., data = BRIEF_AgeOf)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.3537 -2.3268 -0.6309 2.5441 8.6925
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 19.26342 2.13665 9.016 1.14e-14 ***
## SES..3.66. -0.06274 0.02622 -2.393 0.01853 *
## SexMale -0.31561 0.68005 -0.464 0.64355
## AgeMonths -0.02778 0.03029 -0.917 0.36136
## AoAE -0.02023 0.01549 -1.306 0.19443
## Age.of.Language.Exposure..mo. 0.06478 0.02122 3.054 0.00288 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.479 on 103 degrees of freedom
## Multiple R-squared: 0.1552, Adjusted R-squared: 0.1142
## F-statistic: 3.786 on 5 and 103 DF, p-value: 0.003455
anova(plan_all_AoAE, plan_all_AoAE_AoLE)
## Analysis of Variance Table
##
## Model 1: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE
## Model 2: Plan.Organize_RawScore ~ SES..3.66. + Sex + AgeMonths + AoAE +
## Age.of.Language.Exposure..mo.
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 104 1359.5
## 2 103 1246.7 1 112.86 9.3248 0.002877 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Findings: For all subscales, adding Age of Auditory Exposure to the model with demographic variables did not improve model fit (and Age of Auditory Exposure was not a significant predictor of any subsale), but subsequently adding Age of Language exposure to the model with Age of Auditory Exposure and demographic variables did significantly improve model fit (and Age of Language Exposure was a significant predictor of BRIEF-P scores in almost all models).
Possible issue: If this “non-age-appropriate” group of participants is largely from the Early ASL group, and/or has higher SES, our finding that Early exposure to language is the driving factor in EF development may not be valid
Approach: Checking the Age, SES, and Language Group of participants who were “age-appropriate” for the BRIEF-P compared to those who were not
table1::label(BRIEF_AgeOf$AgeMonths) <- "Age (Months)"
table1::label(BRIEF_AgeOf$SES..3.66.) <- "SES"
table1::label(BRIEF_AgeOf$LanguageGroup) <- "Language Group"
table1::table1(~AgeMonths + SES..3.66. + LanguageGroup | Age.Appropriate., data = BRIEF_AgeOf, overall=T)
| no (N=19) |
yes (N=90) |
TRUE (N=109) |
|
|---|---|---|---|
| Age (Months) | |||
| Mean (SD) | 78.7 (5.85) | 55.1 (9.93) | 59.2 (13.0) |
| Median [Min, Max] | 77.0 [72.0, 91.0] | 56.0 [37.0, 71.0] | 59.0 [37.0, 91.0] |
| SES | |||
| Mean (SD) | 47.7 (20.3) | 51.9 (11.6) | 51.2 (13.5) |
| Median [Min, Max] | 58.5 [8.00, 66.0] | 54.0 [15.0, 66.0] | 54.5 [8.00, 66.0] |
| Language Group | |||
| Typically Hearing | 2 (10.5%) | 44 (48.9%) | 46 (42.2%) |
| Early ASL | 4 (21.1%) | 14 (15.6%) | 18 (16.5%) |
| Later English | 3 (15.8%) | 20 (22.2%) | 23 (21.1%) |
| Later ASL | 10 (52.6%) | 12 (13.3%) | 22 (20.2%) |
Participants in the Later groups, specifically the Later ASL group, are over-represented in not having received age-appropriate BRIEF. This should not be a concern for our findings, because if the behaviors surveyed in the BRIEF-P were not appropriate for older kids, then we would not expect the older Later ASL group to have elevated BRIEF scores relative to early-exposed kids–but they do.
Approach Pt 2: Re-run Chi-square with only age-appropriate subset to confirm the non-difference isn’t driven by older Early ASL participants Create dataframe with only age-appropriate participants exposed to language early
BRIEF_early_AgeApp <- subset(BRIEF_AgeApp, BRIEF_AgeApp$Language_Timing=="Early")
BRIEF Scores & Welch two sample t-tests for two “Early” groups within Age-appropriate subset
table1::label(BRIEF_early_AgeApp$GEC_RawScore) <- "Global Executive Composite"
table1::label(BRIEF_early_AgeApp$Inhibit_RawScore) <- "Inhibition"
table1::label(BRIEF_early_AgeApp$Shift_RawScore) <- "Shift"
table1::label(BRIEF_early_AgeApp$Emotional.Control_RawScore) <- "Emotional Control"
table1::label(BRIEF_early_AgeApp$Working.Memory_RawScore) <- "Working Memory"
table1::label(BRIEF_early_AgeApp$Plan.Organize_RawScore) <- "Plan/Organize"
table1(~GEC_RawScore + Inhibit_RawScore + Shift_RawScore + Emotional.Control_RawScore + Working.Memory_RawScore + Plan.Organize_RawScore | Language_Modality, data = BRIEF_early_AgeApp, overall=F)
| English (N=44) |
ASL (N=20) |
|
|---|---|---|
| Global Executive Composite | ||
| Mean (SD) | 88.4 (18.9) | 91.4 (15.9) |
| Median [Min, Max] | 88.0 [63.0, 155] | 91.0 [63.0, 132] |
| Inhibition | ||
| Mean (SD) | 22.8 (6.00) | 24.3 (5.97) |
| Median [Min, Max] | 23.0 [16.0, 46.0] | 22.5 [16.0, 37.0] |
| Shift | ||
| Mean (SD) | 13.2 (3.11) | 14.0 (3.18) |
| Median [Min, Max] | 12.0 [10.0, 24.0] | 13.0 [10.0, 21.0] |
| Emotional Control | ||
| Mean (SD) | 15.1 (3.69) | 15.2 (3.14) |
| Median [Min, Max] | 14.5 [10.0, 24.0] | 14.0 [10.0, 21.0] |
| Working Memory | ||
| Mean (SD) | 22.9 (6.72) | 23.6 (5.00) |
| Median [Min, Max] | 21.0 [17.0, 48.0] | 23.0 [17.0, 37.0] |
| Plan/Organize | ||
| Mean (SD) | 14.4 (3.44) | 14.3 (3.05) |
| Median [Min, Max] | 14.0 [10.0, 23.0] | 14.0 [10.0, 20.0] |
t.test(BRIEF_early_AgeApp$GEC_RawScore~BRIEF_early_AgeApp$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early_AgeApp$GEC_RawScore by BRIEF_early_AgeApp$Language_Modality
## t = -0.65032, df = 43.201, p-value = 0.5189
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -12.152931 6.225658
## sample estimates:
## mean in group English mean in group ASL
## 88.38636 91.35000
t.test(BRIEF_early_AgeApp$Inhibit_RawScore~BRIEF_early_AgeApp$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early_AgeApp$Inhibit_RawScore by BRIEF_early_AgeApp$Language_Modality
## t = -0.90549, df = 37.006, p-value = 0.3711
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -4.724033 1.805851
## sample estimates:
## mean in group English mean in group ASL
## 22.84091 24.30000
t.test(BRIEF_early_AgeApp$Shift_RawScore~BRIEF_early_AgeApp$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early_AgeApp$Shift_RawScore by BRIEF_early_AgeApp$Language_Modality
## t = -0.98755, df = 36.108, p-value = 0.3299
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -2.567680 0.885862
## sample estimates:
## mean in group English mean in group ASL
## 13.15909 14.00000
t.test(BRIEF_early_AgeApp$Emotional.Control_RawScore~BRIEF_early_AgeApp$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early_AgeApp$Emotional.Control_RawScore by BRIEF_early_AgeApp$Language_Modality
## t = -0.096402, df = 42.922, p-value = 0.9237
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.893154 1.720427
## sample estimates:
## mean in group English mean in group ASL
## 15.11364 15.20000
t.test(BRIEF_early_AgeApp$Working.Memory_RawScore~BRIEF_early_AgeApp$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early_AgeApp$Working.Memory_RawScore by BRIEF_early_AgeApp$Language_Modality
## t = -0.45497, df = 48.543, p-value = 0.6512
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -3.718724 2.345996
## sample estimates:
## mean in group English mean in group ASL
## 22.86364 23.55000
t.test(BRIEF_early_AgeApp$Plan.Organize_RawScore~BRIEF_early_AgeApp$Language_Modality)
##
## Welch Two Sample t-test
##
## data: BRIEF_early_AgeApp$Plan.Organize_RawScore by BRIEF_early_AgeApp$Language_Modality
## t = 0.12746, df = 41.286, p-value = 0.8992
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.619017 1.837199
## sample estimates:
## mean in group English mean in group ASL
## 14.40909 14.30000
stargazer(overall_all_base, overall_all_AoLE, overall_all_AoLE_AoAE, overall_all_AoLE_LoAE, overall_all_AoLE_Modality,overall_all_AoLE_HearStat, type= "html", title = "Linear Regression Results", align=TRUE, dep.var.labels=c("Global Executive Composite Raw Scores"), covariate.labels=c("Socioeconomic Status (SES)", "Sex (Male)", "Age (Months)", "Age of Exposure to Language (Months)", "Age of Auditory Exposure (Months)", "Length of Auditory Experience (Months)", "Language Modality (ASL)", "Hearing Status (Deaf)", "Hearing Status (HoH)"), out="BRIEF_supp_table1.html")
stargazer(inhibition_all_base, inhibition_all_AoLE, inhibition_all_AoLE_AoAE, inhibition_all_AoLE_LoAE, inhibition_all_AoLE_Modality, inhibition_all_AoLE_HearStat, type= "html", title = "Linear Regression Results", align=TRUE, dep.var.labels=c("Inhibition Raw Scores"), covariate.labels=c("Socioeconomic Status (SES)", "Sex (Male)", "Age (Months)", "Age of Exposure to Language (Months)", "Age of Auditory Exposure (Months)", "Length of Auditory Experience (Months)", "Language Modality (ASL)", "Hearing Status (Deaf)", "Hearing Status (HoH)"), out="BRIEF_supp_table2.html")
stargazer(shift_all_base, shift_all_AoLE, shift_all_AoLE_AoAE, shift_all_AoLE_LoAE, shift_all_AoLE_Modality, shift_all_AoLE_HearStat, type= "html", title = "Linear Regression Results", align=TRUE, dep.var.labels=c("Shift Raw Scores"), covariate.labels=c("Socioeconomic Status (SES)", "Sex (Male)", "Age (Months)", "Age of Exposure to Language (Months)", "Age of Auditory Exposure (Months)", "Length of Auditory Experience (Months)", "Language Modality (ASL)", "Hearing Status (Deaf)", "Hearing Status (HoH)"), out="BRIEF_supp_table3.html")
stargazer(emotctrl_all_base, emotctrl_all_AoLE, emotctrl_all_AoLE_AoAE, emotctrl_all_AoLE_LoAE, emotctrl_all_AoLE_Modality, emotctrl_all_AoLE_HearStat, type= "html", title = "Linear Regression Results", align=TRUE, dep.var.labels=c("Emotional Control Raw Scores"), covariate.labels=c("Socioeconomic Status (SES)", "Sex (Male)", "Age (Months)", "Age of Exposure to Language (Months)", "Age of Auditory Exposure (Months)", "Length of Auditory Experience (Months)", "Language Modality (ASL)", "Hearing Status (Deaf)", "Hearing Status (HoH)"), out="BRIEF_supp_table4.html")
stargazer(wm_all_base, wm_all_AoLE, wm_all_AoLE_AoAE, wm_all_AoLE_LoAE, wm_all_AoLE_Modality,wm_all_AoLE_HearStat, type= "html", title = "Linear Regression Results", align=TRUE, dep.var.labels=c("Working Memory Raw Scores"), covariate.labels=c("Socioeconomic Status (SES)", "Sex (Male)", "Age (Months)", "Age of Exposure to Language (Months)", "Age of Auditory Exposure (Months)", "Length of Auditory Experience (Months)", "Language Modality (ASL)", "Hearing Status (Deaf)", "Hearing Status (HoH)"), out="BRIEF_supp_table5.html")
stargazer(plan_all_base, plan_all_AoLE, plan_all_AoLE_AoAE, plan_all_AoLE_LoAE, plan_all_AoLE_Modality, plan_all_AoLE_HearStat, type= "html", title = "Linear Regression Results", align=TRUE, dep.var.labels=c("Planning/Organization Raw Scores"), covariate.labels=c("Socioeconomic Status (SES)", "Sex (Male)", "Age (Months)", "Age of Exposure to Language (Months)", "Age of Auditory Exposure (Months)", "Length of Auditory Experience (Months)", "Language Modality (ASL)", "Hearing Status (Deaf)", "Hearing Status (HoH)"), out="BRIEF_supp_table6.html")
table1::label(BRIEF_AgeApp$AgeMonths) <- "Age (Months)"
table1::label(BRIEF_AgeApp$Sex) <- "Sex"
table1::label(BRIEF_AgeApp$Race_recoded) <- "Race"
table1::label(BRIEF_AgeApp$Ethnicity_recoded) <- "Ethnicity"
table1::label(BRIEF_AgeApp$SES..3.66.) <- "SES"
table1::table1(~AgeMonths + Sex + Race_recoded + Ethnicity_recoded + SES..3.66. | LanguageGroup, data = BRIEF_AgeApp, overall=F)
| Typically Hearing (N=44) |
Early ASL (N=20) |
Later English (N=20) |
Later ASL (N=14) |
|
|---|---|---|---|---|
| Age (Months) | ||||
| Mean (SD) | 53.7 (9.80) | 54.5 (8.98) | 58.3 (9.67) | 56.8 (11.3) |
| Median [Min, Max] | 54.0 [37.0, 71.0] | 54.5 [41.0, 71.0] | 60.0 [37.0, 71.0] | 60.0 [37.0, 71.0] |
| Sex | ||||
| Female | 23 (52.3%) | 11 (55.0%) | 10 (50.0%) | 7 (50.0%) |
| Male | 21 (47.7%) | 9 (45.0%) | 10 (50.0%) | 7 (50.0%) |
| Race | ||||
| Asian | 0 (0%) | 0 (0%) | 0 (0%) | 1 (7.1%) |
| Black or African American | 0 (0%) | 0 (0%) | 1 (5.0%) | 0 (0%) |
| White | 41 (93.2%) | 18 (90.0%) | 17 (85.0%) | 11 (78.6%) |
| More than one | 3 (6.8%) | 1 (5.0%) | 2 (10.0%) | 1 (7.1%) |
| Other/Missing | 0 (0%) | 1 (5.0%) | 0 (0%) | 1 (7.1%) |
| Ethnicity | ||||
| Hispanic | 3 (6.8%) | 0 (0%) | 1 (5.0%) | 2 (14.3%) |
| Non-Hispanic | 40 (90.9%) | 13 (65.0%) | 15 (75.0%) | 10 (71.4%) |
| Prefer not to answer | 0 (0%) | 1 (5.0%) | 0 (0%) | 0 (0%) |
| Missing | 1 (2.3%) | 6 (30.0%) | 4 (20.0%) | 2 (14.3%) |
| SES | ||||
| Mean (SD) | 55.5 (8.83) | 47.0 (16.0) | 49.6 (11.3) | 46.7 (14.2) |
| Median [Min, Max] | 56.0 [21.5, 66.0] | 53.8 [19.0, 66.0] | 50.5 [15.0, 66.0] | 50.8 [9.00, 62.0] |
BRIEF_AgeApp_GEC_over60 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Global Executive Composite T-score < 60 (n)" = sum(GEC_Tscore < 60), "Global Executive Composite T-score > 60 (n)" = sum(GEC_Tscore >= 60))
BRIEF_AgeApp_GEC_over60 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Global Executive Composite T-score < 60 (n) | Global Executive Composite T-score > 60 (n) |
|---|---|---|
| Typically Hearing | 39 | 5 |
| Early ASL | 17 | 3 |
| Later English | 14 | 6 |
| Later ASL | 10 | 4 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
GEC_60rr <- matrix(c(39,5,17,3, 14, 6, 10, 4), 4, 2, byrow=TRUE)
dimnames(GEC_60rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "GEC T-score" = c("Not Elevated", "Elevated"))
riskratio.small(GEC_60rr)
## $data
## GEC T-score
## Group Not Elevated Elevated Total
## Typically Hearing 39 5 44
## Early ASL 17 3 20
## Later English 14 6 20
## Later ASL 10 4 14
## Total 80 18 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.000000 NA NA
## Early ASL 1.125000 0.2974811 4.254472
## Later English 2.250000 0.7774878 6.511356
## Later ASL 2.142857 0.6656236 6.898548
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.68785362 0.69697927 0.68348090
## Later English 0.08984932 0.08354265 0.06699369
## Later ASL 0.16123264 0.19823228 0.12141705
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
#DHH groups not significantly more likely to have elevated GEC T-scores relative to TH group, but the two later groups are close
BRIEF_AgeApp_Inhibit_over60 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Inhibit T-score < 60 (n)" = sum(Inhibit_Tscore < 60), "Inhibit T-score > 60 (n)" = sum(Inhibit_Tscore >= 60))
BRIEF_AgeApp_Inhibit_over60 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Inhibit T-score < 60 (n) | Inhibit T-score > 60 (n) |
|---|---|---|
| Typically Hearing | 39 | 5 |
| Early ASL | 16 | 4 |
| Later English | 17 | 3 |
| Later ASL | 10 | 4 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
Inhibit_60rr <- matrix(c(39,5,16,4, 17, 3, 10, 4), 4, 2, byrow=TRUE)
dimnames(Inhibit_60rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "Inhibit T-score" = c("Not Elevated", "Elevated"))
riskratio.small(Inhibit_60rr)
## $data
## Inhibit T-score
## Group Not Elevated Elevated Total
## Typically Hearing 39 5 44
## Early ASL 16 4 20
## Later English 17 3 20
## Later ASL 10 4 14
## Total 82 16 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.000000 NA NA
## Early ASL 1.500000 0.4500510 4.999433
## Later English 1.125000 0.2974811 4.254472
## Later ASL 2.142857 0.6656236 6.898548
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.3869141 0.4434313 0.3569397
## Later English 0.6878536 0.6969793 0.6834809
## Later ASL 0.1612326 0.1982323 0.1214171
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
#DHH groups *not* significantly more likely to have elevated Inhibit T-scores relative to TH group, but the Later ASL group close
BRIEF_AgeApp_Shift_over60 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Shift T-score < 60 (n)" = sum(Shift_Tscore < 60), "Shift T-score > 60 (n)" = sum(Shift_Tscore >= 60))
BRIEF_AgeApp_Shift_over60 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Shift T-score < 60 (n) | Shift T-score > 60 (n) |
|---|---|---|
| Typically Hearing | 42 | 2 |
| Early ASL | 18 | 2 |
| Later English | 18 | 2 |
| Later ASL | 10 | 4 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
Shift_60rr <- matrix(c(42,2,18,2, 18, 2, 10, 4), 4, 2, byrow=TRUE)
dimnames(Shift_60rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "Shift T-score" = c("Not Elevated", "Elevated"))
riskratio.small(Shift_60rr)
## $data
## Shift T-score
## Group Not Elevated Elevated Total
## Typically Hearing 42 2 44
## Early ASL 18 2 20
## Later English 18 2 20
## Later ASL 10 4 14
## Total 88 10 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.000000 NA NA
## Early ASL 1.500000 0.2272096 9.90275
## Later English 1.500000 0.2272096 9.90275
## Later ASL 4.285714 0.8763638 20.95859
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.45602919 0.58311299 0.4033953
## Later English 0.45602919 0.58311299 0.4033953
## Later ASL 0.02789668 0.02564615 0.0101395
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
#Later ASL group significantly more likely to have elevated Shift T-scores relative to TH group (3.75 times more likely) according to p-values (even though risk ratio CI does include 1)
BRIEF_AgeApp_EmoCtrl_over60 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Emotional Control T-score < 60 (n)" = sum(Emotional.Control_Tscore < 60), "Emotional Control T-score > 60 (n)" = sum(Emotional.Control_Tscore >= 60))
BRIEF_AgeApp_EmoCtrl_over60 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Emotional Control T-score < 60 (n) | Emotional Control T-score > 60 (n) |
|---|---|---|
| Typically Hearing | 33 | 11 |
| Early ASL | 18 | 2 |
| Later English | 18 | 2 |
| Later ASL | 11 | 3 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
EmoCtrl_60rr <- matrix(c(33,11, 18,2, 18,2, 11,3), 4, 2, byrow=TRUE)
dimnames(EmoCtrl_60rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "Emotional Control T-score" = c("Not Elevated", "Elevated"))
riskratio.small(EmoCtrl_60rr)
## $data
## Emotional Control T-score
## Group Not Elevated Elevated Total
## Typically Hearing 33 11 44
## Early ASL 18 2 20
## Later English 18 2 20
## Later ASL 11 3 14
## Total 80 18 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.0000000 NA NA
## Early ASL 0.3750000 0.09147347 1.537331
## Later English 0.3750000 0.09147347 1.537331
## Later ASL 0.8035714 0.26060473 2.477802
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.1830447 0.200987 0.1668176
## Later English 0.1830447 0.200987 0.1668176
## Later ASL 0.8196528 1.000000 0.7856283
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
#Later English borderline *less* likely than TH to have elevated Emotional Control T-scores (but there do seem to be an unexpectedly high number of TH kids in the "elevated" score range, and Later English group only on average 4 months older than TH kids)
BRIEF_AgeApp_WorkMem_over60 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Working Memory T-score < 60 (n)" = sum(Working.Memory_Tscore < 60), "Working Memory T-score > 60 (n)" = sum(Working.Memory_Tscore >= 60))
BRIEF_AgeApp_WorkMem_over60 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Working Memory T-score < 60 (n) | Working Memory T-score > 60 (n) |
|---|---|---|
| Typically Hearing | 38 | 6 |
| Early ASL | 16 | 4 |
| Later English | 15 | 5 |
| Later ASL | 8 | 6 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
WorkMem_60rr <- matrix(c(38,6,16,4,15,5,8,6), 4, 2, byrow=TRUE)
dimnames(WorkMem_60rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "Working Memory T-score" = c("Not Elevated", "Elevated"))
riskratio.small(WorkMem_60rr)
## $data
## Working Memory T-score
## Group Not Elevated Elevated Total
## Typically Hearing 38 6 44
## Early ASL 16 4 20
## Later English 15 5 20
## Later ASL 8 6 14
## Total 77 21 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.000000 NA NA
## Early ASL 1.285714 0.4073295 4.058290
## Later English 1.607143 0.5553484 4.650969
## Later ASL 2.755102 1.0564549 7.184961
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.53280902 0.71159622 0.51576333
## Later English 0.29363909 0.29746086 0.26403699
## Later ASL 0.03310511 0.05208747 0.01873012
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
#Later ASL significantly more likely to have elevated Working Memory T-scores relative to TH group
BRIEF_AgeApp_PlanOrg_over60 <- BRIEF_AgeApp %>%
group_by(LanguageGroup) %>%
summarise("Plan/Organize T-score < 60 (n)" = sum(Plan.Organize_Tscore < 60), "Plan/Organize T-score > 60 (n)" = sum(Plan.Organize_Tscore >= 60))
BRIEF_AgeApp_PlanOrg_over60 %>%
kable() %>%
kable_styling(bootstrap_options = "striped")
| LanguageGroup | Plan/Organize T-score < 60 (n) | Plan/Organize T-score > 60 (n) |
|---|---|---|
| Typically Hearing | 37 | 7 |
| Early ASL | 17 | 3 |
| Later English | 14 | 6 |
| Later ASL | 10 | 4 |
#The numbers in this matrix are static generated from the code above. If the numbers in the matrix change, the code also needs to be altered
PlanOrg_60rr <- matrix(c(37,7,17,3, 14, 6, 10, 4), 4, 2, byrow=TRUE)
dimnames(PlanOrg_60rr) <- list("Group" = c("Typically Hearing", "Early ASL", "Later English", "Later ASL"), "Plan/Organize T-score" = c("Not Elevated", "Elevated"))
riskratio.small(PlanOrg_60rr)
## $data
## Plan/Organize T-score
## Group Not Elevated Elevated Total
## Typically Hearing 37 7 44
## Early ASL 17 3 20
## Later English 14 6 20
## Later ASL 10 4 14
## Total 78 20 98
##
## $measure
## risk ratio with 95% C.I.
## Group estimate lower upper
## Typically Hearing 1.000000 NA NA
## Early ASL 0.843750 0.2429636 2.930127
## Later English 1.687500 0.6501810 4.379790
## Later ASL 1.607143 0.5506079 4.691012
##
## $p.value
## two-sided
## Group midp.exact fisher.exact chi.square
## Typically Hearing NA NA NA
## Early ASL 0.9529974 1.0000000 0.9260295
## Later English 0.2210706 0.3140510 0.1940420
## Later ASL 0.3257076 0.4333345 0.2925137
##
## $correction
## [1] FALSE
##
## attr(,"method")
## [1] "small sample-adjusted UMLE & normal approx (Wald) CI"
#DHH groups *not* significantly more likely to have elevated Inhibit T-scores relative to TH group